CATALOGUE·EVERY LIVE CORPUS
THE
CATALOGUE
559 live corpora, one row each, pinned and licence-checked, plus 12,642 corpora contributed by their authors. Agents read the same list as text in llms.txt.
Live corpora · 559
| corpus | source | domain | licence | size | G7 | for |
|---|---|---|---|---|---|---|
| 12-factor-agents | humanlayer/12-factor-agents | AI | Apache-2.0 | 44.9 MB | not yet graded | an agent builder queries this for the reliability and design principles of production LLM agents, the twelve factors |
| 18f-content-guide | 18F/content-guide | WRITE | CC0-1.0 | 416.9 KB | not yet graded | an agent writing plain-language content queries this for active-voice, concise-heading and inclusive-wording rules |
| a11y-project-content | a11yproject/a11yproject.com | WRITE | Apache-2.0 | 22.3 MB | not yet graded | An agent writing accessible UI copy and alt text queries this for The A11Y Project's accessibility checklist and plain-language accessibility articles. |
| accord-legal-models | accordproject/models | LAW | Apache-2.0 | 6.7 MB | not yet graded | law |
| act | nektos/act | OPS | MIT | 108.6 MB | not yet graded | An agent debugging or running GitHub Actions locally queries this for workflow-runner emulation and event/context handling. |
| ada-2010 | xerj-org/corpus-ada-2010 | LAW | NONE-FOUND | 556.4 KB | G7 5.0/5 | ADA 2010 Standards for Accessible Design: exact dimensional and technical requirements |
| adalflow | SylphAI-Inc/AdalFlow | AI | MIT | 41.4 MB | not yet graded | An agent building a trainable LLM pipeline queries this for prompt/parameter optimization and retriever-generator composition. |
| admission-compare | Thneoly/r2r-jev | AI | Apache-2.0 | 146.9 KB | not yet graded | an agent comparing evidence admission with a direct threshold queries this for how admission-compare runs the two-arm experiment over CSV judgment traces |
| adyen-openapi | Adyen/adyen-openapi | API | MIT | 54.0 MB | not yet graded | An agent modeling payment APIs queries this for Adyen's OpenAPI definitions and payment-domain conventions. |
| aeron | real-logic/aeron | NET | Apache-2.0 | 20.2 MB | not yet graded | an agent working on ultra-low-latency UDP/IPC messaging queries this for its log-buffer transport, flow control, and term-based framing |
| age | FiloSottile/age | CRYPTO | BSD-3-Clause | 684.2 KB | not yet graded | An agent working on a file-encryption tool or the age format queries this for the reference X25519 + ChaCha20-Poly1305 recipient/identity design and stanza form |
| agent-session-trajectories | chunxiaoxx/nautilus-compass | DB | CC-BY-4.0 | 0 B | not yet graded | an agent retrieving from its own or another agent's multi-session history: raw trajectories with timestamps, failure turns kept verbatim, so a hit that says you |
| agentbench | THUDM/AgentBench | AI | Apache-2.0 | 43.1 MB | not yet graded | An agent benchmarking agentic abilities queries this for multi-environment agent evaluation tasks. |
| agentic-security-ietf-corpus | drk-ram/agentic-security-ietf-corpus | SEC | Apache-2.0 | 151.7 KB | not yet graded | an agent building MCP/A2A security gates, scope guards, or trace verifiers queries this for clause-level IETF Internet-Drafts (RFC 8785 JCS trace verification, |
| agentops | AgentOps-AI/agentops | AI | MIT | 166.5 MB | not yet graded | An agent monitoring multi-step agent runs queries this for session replay, cost tracking, and per-step event instrumentation. |
| agentscope | modelscope/agentscope | AI | Apache-2.0 | 16.3 MB | not yet graded | An agent needing a robust, distributed multi-agent runtime queries this for message-passing and fault-tolerant orchestration. |
| agno | agno-agi/agno | AI | Apache-2.0 | 58.6 MB | not yet graded | An agent builder wanting fast, model-agnostic agents queries this for multimodal agent plus tool/memory primitives. |
| aider | Aider-AI/aider | AI | Apache-2.0 | 74.1 MB | not yet graded | An agent editing a real git repo queries this for repo-map building, diff-based edits, and commit workflows. |
| aiohttp | aio-libs/aiohttp | NET | Apache-2.0 | 5.1 MB | not yet graded | an agent working on an async HTTP client/server stack queries this for streaming bodies, websocket upgrades, and connection reuse |
| aioquic | aiortc/aioquic | NET | BSD-3-Clause | 1.1 MB | not yet graded | An agent learning QUIC/HTTP3 internals queries this for a readable implementation of packet parsing, the handshake, and QPACK header compression. |
| alertmanager | prometheus/alertmanager | OPS | Apache-2.0 | 3.9 MB | not yet graded | An agent routing alerts queries this for Alertmanager grouping, inhibition, silencing and receiver/notification config. |
| alex-inclusive-language | get-alex/alex | WRITE | MIT | 479.2 KB | not yet graded | An agent checking copy for insensitive or non-inclusive wording queries this for the terms to flag and their neutral replacements. |
| ann-benchmarks | erikbern/ann-benchmarks | DB | MIT | 2.6 MB | not yet graded | An agent working on measuring ANN recall-vs-latency and comparing vector indexes queries this for ann-benchmarks' datasets, recall methodology and runner harnes |
| annoy | spotify/annoy | DB | Apache-2.0 | 824.4 KB | not yet graded | an agent working on read-mostly ANN queries this for the forest-of-trees build and on-disk format |
| anthropic-cookbook | anthropics/anthropic-cookbook | AI | MIT | 207.8 MB | not yet graded | An agent learning effective tool-use and RAG recipes queries this for runnable notebook patterns. |
| anything-llm | Mintplex-Labs/anything-llm | AI | MIT | 56.6 MB | not yet graded | An agent packaged as a desktop/server RAG app queries this for workspace, embed, and connector patterns. |
| api-blueprint | apiaryio/api-blueprint | API | MIT | 407.9 KB | not yet graded | An agent designing a human-readable API description format queries this for the API Blueprint markdown resource/action grammar and MSON. |
| api-common-protos | googleapis/api-common-protos | API | Apache-2.0 | 339.4 KB | not yet graded | An agent adding HTTP/gRPC annotations, field_behavior or resource names to protobuf APIs queries this for Google's common API proto types (google.api.*). |
| api-linter | googleapis/api-linter | API | Apache-2.0 | 2.7 MB | not yet graded | An agent enforcing resource-oriented API conventions queries this for how AIP rules are implemented as lint checks over protobuf. |
| arazzo-spec | OAI/Arazzo-Specification | API | Apache-2.0 | 5.2 MB | not yet graded | An agent orchestrating multi-call API workflows queries this for the Arazzo Specification, which describes deterministic sequences and dependencies across OpenA |
| argo-cd | argoproj/argo-cd | OPS | Apache-2.0 | 214.7 MB | not yet graded | An agent building GitOps delivery queries this for declarative continuous deployment, sync waves/hooks and drift detection. |
| argo-workflows | argoproj/argo-workflows | OPS | Apache-2.0 | 79.6 MB | not yet graded | An agent orchestrating pipelines on Kubernetes queries this for DAG/step workflow specs, artifacts, retries and the workflow controller. |
| aria-practices | w3c/aria-practices | WEB | W3C-20150513 | 8.7 MB | not yet graded | An agent building an accessible widget queries this ARIA Authoring Practices Guide for the canonical keyboard-interaction and ARIA patterns per component. |
| arkworks-algebra | arkworks-rs/algebra | CRYPTO | Apache-2.0 | 2.6 MB | not yet graded | An agent building ZK proof systems queries this for arkworks' finite-field, elliptic-curve and polynomial arithmetic traits. |
| arrow-rs | apache/arrow-rs | DB | Apache-2.0 | 20.1 MB | not yet graded | an agent working on columnar memory queries this for the Rust arrow arrays, kernels and IPC paths |
| arroyo | ArroyoSystems/arroyo | DB | Apache-2.0 | 20.6 MB | not yet graded | an agent working on SQL stream processing with windowing queries this for a Rust engine's state backends and exactly-once checkpoints |
| astro | withastro/astro | WEB | MIT | 49.2 MB | not yet graded | An agent building a content/SSG framework queries this for Astro's islands architecture, partial hydration and content collections. |
| astro-docs | withastro/docs | WEB | MIT | 76.5 MB | not yet graded | An agent building content-driven sites with islands/SSR queries this for the Astro documentation. |
| asyncapi-bindings | asyncapi/bindings | API | Apache-2.0 | 574.5 KB | not yet graded | An agent describing protocol-specific channel/operation details in AsyncAPI queries this for the per-protocol binding schemas. |
| asyncapi-spec | asyncapi/spec | API | Apache-2.0 | 1.7 MB | not yet graded | An agent working on event-driven or message-based API definitions queries this for the AsyncAPI specification's document structure, protocol bindings (Kafka/MQT |
| asyncapi-spec-json-schemas | asyncapi/spec-json-schemas | API | Apache-2.0 | 4.8 MB | not yet graded | An agent machine-validating AsyncAPI documents queries this for the JSON Schema definitions of every AsyncAPI version. |
| asyncapi-studio | asyncapi/studio | API | Apache-2.0 | 5.5 MB | not yet graded | An agent building a visual editor/validator for AsyncAPI documents queries this for Studio's parsing, validation and example documents. |
| atoma-run--atoma | atoma-run/atoma | AI | AGPL-3.0-only | 130.3 MB | not yet graded | an agent building a multi-tier LLM orchestrator queries this for how atoma runs one plan/validate/execute/validate supervision loop across three agent ranks, ke |
| atomic-agents | BrainBlend-AI/atomic-agents | AI | MIT | 10.9 MB | not yet graded | An agent composed from small reusable blocks queries this for a schema-driven, modular agent architecture. |
| attack-flow | center-for-threat-informed-defense/attack-flow | SEC | Apache-2.0 | 36.2 MB | not yet graded | security |
| attack-mappings-explorer | center-for-threat-informed-defense/mappings-explorer | SEC | Apache-2.0 | 60.2 MB | not yet graded | An agent correlating controls to adversary behavior queries this for CTID mappings of ATT&CK techniques to NIST 800-53, CIS, Azure/AWS/GCP security controls. |
| autogen | microsoft/autogen | AI | CC-BY-4.0 | 46.1 MB | not yet graded | An agent built from conversational multi-agent groups queries this for event-driven agent-to-agent messaging patterns. |
| autorag | Marker-Inc-Korea/AutoRAG | AI | Apache-2.0 | 82.4 MB | not yet graded | An agent tuning a RAG pipeline queries this for automated evaluation/optimization across retriever+generator component combinations. |
| avro | apache/avro | NET | Apache-2.0 | 17.1 MB | not yet graded | NET |
| awesome-ai-system-prompts | dontriskit/awesome-ai-system-prompts | AI | MIT | 1.2 MB | not yet graded | an agent designing a system prompt queries this for how frontier tools (ChatGPT, Claude, Cursor, v0) are actually system-prompted, with annotated patterns |
| awesome-llm-apps | Shubhamsaboo/awesome-llm-apps | AI | Apache-2.0 | 85.9 MB | not yet graded | An agent looking for end-to-end example apps queries this for a large catalog of RAG/agent app implementations with code. |
| awesome-mcp-servers | punkpeye/awesome-mcp-servers | AI | MIT | 2.3 MB | not yet graded | an agent wiring up MCP queries this for which server exists for a capability or integration and where to get it |
| awesome-sre | dastergon/awesome-sre | OPS | CC0-1.0 | 91.6 KB | not yet graded | An agent researching reliability tooling and practice queries this curated index for pointers to SRE, observability, and production-engineering resources. |
| axe-core | dequelabs/axe-core | WEB | MPL-2.0 | 44.2 MB | not yet graded | An agent writing automated accessibility checks queries this for axe's rule set and how each WCAG/ARIA violation is actually detected in the DOM. |
| axolotl | axolotl-ai-cloud/axolotl | AI | Apache-2.0 | 16.4 MB | not yet graded | An agent fine-tuning open LLMs queries this for YAML-config SFT/LoRA/QLoRA training recipes. |
| badger | dgraph-io/badger | DB | Apache-2.0 | 1.3 MB | not yet graded | an agent working on pure-Go KV stores queries this for its LSM-with-value-log design and transaction implementation |
| bandit-plugins | PyCQA/bandit | SEC | Apache-2.0 | 4.8 MB | not yet graded | An agent writing a Python SAST or secure-coding linter queries this for Bandit's plugin checks (AST patterns for injection, weak crypto, insecure deserializatio |
| bebop | betwixt-labs/bebop | NET | Apache-2.0 | 7.7 MB | not yet graded | an agent working on a schema-based binary wire format queries this for compiled encoders/decoders and compact record layout |
| biome | biomejs/biome | WEB | Apache-2.0 | 158.4 MB | not yet graded | A Rust agent building a formatter/linter queries this for Biome's Rust lossless syntax tree (Rowan-style), formatter IR and lint-rule engine. |
| blake2 | BLAKE2/BLAKE2 | CRYPTO | CC0-1.0 | 5.0 MB | not yet graded | an agent working on a BLAKE2b/2s/2x hash or keyed MAC queries this for the reference and SSE/AVX-optimized implementations |
| blake3 | BLAKE3-team/BLAKE3 | CRYPTO | Apache-2.0 | 1.7 MB | not yet graded | An agent implementing fast cryptographic hashing or content-addressed digests queries this for the BLAKE3 tree-hash construction, SIMD lanes, and keyed/derive-k |
| bleve | blevesearch/bleve | DB | Apache-2.0 | 8.4 MB | not yet graded | An agent working on an embeddable full-text index with pluggable storage queries this for Bleve's analysis pipeline, index mapping and scorch segment format in |
| blst | supranational/blst | CRYPTO | Apache-2.0 | 3.7 MB | not yet graded | An agent implementing BLS12-381 signatures queries this for supranational's assembly-optimised pairings and signature aggregation. |
| bluge | blugelabs/bluge | DB | Apache-2.0 | 1.8 MB | not yet graded | An agent working on a segment-based full-text library decoupled from document storage queries this for Bluge's reader/writer and segment design. |
| bookkeeper | apache/bookkeeper | DB | Apache-2.0 | 22.6 MB | not yet graded | An agent working on a replicated write-ahead log / ledger storage layer queries this for BookKeeper's ledger, ensemble and journal/entry-log design. |
| boringtun | cloudflare/boringtun | CRYPTO | BSD-3-Clause | 463.1 KB | G7 3.0/5 | an agent implementing a VPN data plane queries this for the wireguard protocol implementation and its async runtime use |
| borsh-rs | near/borsh-rs | NET | Apache-2.0 | 995.1 KB | not yet graded | An agent needing deterministic binary serialization queries this for a canonical, spec-defined byte layout suited to hashing and consensus. |
| botan | randombit/botan | CRYPTO | BSD-2-Clause | 45.8 MB | not yet graded | An agent needing a broad C++ crypto toolkit queries this for Botan's cipher/public-key/TLS implementations and constant-time utilities. |
| boulder | letsencrypt/boulder | CRYPTO | MPL-2.0 | 39.1 MB | not yet graded | an agent working on an ACME certificate authority queries this for Let's Encrypt's issuance pipeline, validation and rate-limit design |
| box-openapi | box/box-openapi | API | Apache-2.0 | 3.8 MB | not yet graded | An agent modelling a content/file-management REST API with explicit API versioning queries this for Box's versioned OpenAPI 3.0 definition. |
| brex-prompt-engineering | brexhq/prompt-engineering | AI | MIT | 71.8 KB | not yet graded | An agent writing or refining LLM prompts queries this for Brex's battle-tested prompt-engineering patterns, anti-patterns, and reasoning techniques in plain pro |
| brotli | google/brotli | NET | MIT | 63.4 MB | not yet graded | an agent working on web content encoding queries this for the static dictionary and transform handling |
| brpc | apache/brpc | NET | Apache-2.0 | 39.9 MB | not yet graded | an agent working on an industrial C++ RPC framework queries this for multi-protocol server design, load balancing, and bthread concurrency |
| bson-rust | mongodb/bson-rust | NET | MIT | 1.3 MB | not yet graded | An agent implementing BSON queries this for document encoding, type markers, and the length-prefixed field layout. |
| btrblocks | maxi-k/btrblocks | DB | MIT | 1.4 MB | not yet graded | an agent working on columnar compression schemes queries this for the BtrBlocks cascading-encoding research implementation |
| buf | bufbuild/buf | API | Apache-2.0 | 13.0 MB | not yet graded | An agent managing protobuf schemas (lint, breaking-change detection, build, deps) queries this for Buf's module model and rule sets. |
| buildpacks-pack | buildpacks/pack | OPS | Apache-2.0 | 89.1 MB | not yet graded | An agent building images without Dockerfiles queries this for the Cloud Native Buildpacks lifecycle and the pack CLI. |
| bulletproofs | dalek-cryptography/bulletproofs | CRYPTO | MIT | 1.5 MB | not yet graded | An agent implementing range proofs or inner-product arguments queries this for the dalek Bulletproofs implementation. |
| c-ares | c-ares/c-ares | NET | MIT | 3.6 MB | not yet graded | NET |
| caddy | caddyserver/caddy | NET | Apache-2.0 | 4.3 MB | not yet graded | an agent building a TLS-serving reverse proxy queries this for cert automation and config-reload design |
| calcite | apache/calcite | DB | Apache-2.0 | 34.5 MB | not yet graded | An agent working on SQL parsing, relational algebra and cost-based optimization queries this for Calcite's RelNode algebra, planner rules and optimizer framewor |
| caldera | mitre/caldera | SEC | Apache-2.0 | 7.5 MB | not yet graded | an agent automating red-team operations queries this for MITRE Caldera ability definitions, adversary profiles and planners. |
| caldera-stockpile | mitre/stockpile | SEC | Apache-2.0 | 10.8 MB | not yet graded | An agent modeling automated adversary behavior queries this for MITRE CALDERA abilities, adversary profiles and planners (ATT&CK-mapped command definitions). |
| caliber-ehr-phenotypes | spiros/chronological-map-phenotypes | HEALTH | MIT | 3.7 MB | not yet graded | health |
| camel | camel-ai/camel | AI | Apache-2.0 | 144.9 MB | not yet graded | An agent studying role-playing multi-agent societies queries this for cooperative agent communication patterns. |
| candle | huggingface/candle | AI | Apache-2.0 | 15.0 MB | not yet graded | A Rust engine adding embeddings/model inference queries this for a minimalist Rust tensor/ML framework and model implementations. |
| caniuse | Fyrd/caniuse | WEB | CC-BY-4.0 | 23.2 MB | not yet graded | an agent targeting cross-browser features queries this for per-feature browser-version support data, notes, and known bugs. |
| cap-examples | harvard-lil/cap-examples | LAW | MIT | 6.0 MB | not yet graded | An agent working with U.S. case-law bulk data queries this for worked examples (bulk extract, full-text search, ngrams) of using the Harvard Caselaw Access Proj |
| capa-rules | mandiant/capa-rules | SEC | Apache-2.0 | 904.3 KB | not yet graded | an agent triaging a binary queries this for capa's rules mapping code features to malware capabilities and ATT&CK techniques. |
| capnproto | capnproto/capnproto | NET | MIT | 10.1 MB | not yet graded | an agent doing capability RPC queries this for the serialisation layout, promise pipelining and the security hardening history |
| carbon | carbon-design-system/carbon | WEB | Apache-2.0 | 47.0 MB | not yet graded | An agent building against IBM's design system queries this for Carbon component APIs, design tokens and accessibility/content guidance. |
| cassandra | apache/cassandra | DB | Apache-2.0 | 136.9 MB | not yet graded | An agent working on LSM wide-column storage, SSTables and compaction strategies queries this for Cassandra's storage-engine and gossip distribution model. |
| catala-law-lang | CatalaLang/catala | LAW | Apache-2.0 | 5.9 MB | not yet graded | An agent translating statutes/tax law into executable, auditable code queries this for a law-oriented DSL and worked legislative examples. |
| cdc-clinical | xerj-org/corpus-cdc-clinical | HEALTH | US-government public domain (CDC) | 1.1 MB | G7 5.0/5 | a clinician or medical writer queries this for the current CDC-recommended STI regimen and dose, and measles clinical signs |
| cdk8s | cdk8s-team/cdk8s | OPS | Apache-2.0 | 55.4 MB | not yet graded | An agent generating Kubernetes manifests from code queries this for construct/synthesis patterns and typed-API usage. |
| cel-spec | google/cel-spec | API | Apache-2.0 | 1.4 MB | not yet graded | An agent adding an expression language for API policy/validation queries this for the Common Expression Language grammar and semantics. |
| cell-ontology | obophenotype/cell-ontology | HEALTH | CC-BY-4.0 | 214.5 MB | not yet graded | An agent labeling cell types in pathology/histology/immunology pipelines queries this for the Cell Ontology (CL) of in-vivo cell types. |
| cereal | USCiLab/cereal | NET | BSD-3-Clause | 2.1 MB | not yet graded | an agent working on a header-only C++ serialization library queries this for archive abstractions over binary/JSON/XML |
| cert-manager | cert-manager/cert-manager | CRYPTO | Apache-2.0 | 14.7 MB | not yet graded | An agent automating certificates in Kubernetes queries this for cert-manager's issuer/certificate CRDs and ACME controller. |
| certmagic | caddyserver/certmagic | CRYPTO | Apache-2.0 | 646.2 KB | not yet graded | an agent working on automatic HTTPS in a Go service queries this for CertMagic's on-demand issuance, renewal and storage abstraction |
| certstrap | square/certstrap | CRYPTO | Apache-2.0 | 207.1 KB | not yet graded | An agent bootstrapping a simple internal CA queries this for Square's certstrap CA/certificate generation CLI. |
| cfn-nag | stelligent/cfn_nag | SEC | MIT | 1.3 MB | not yet graded | an agent reviewing CloudFormation templates queries this for cfn_nag's Ruby rule definitions that flag insecure AWS resource configuration. |
| cfssl | cloudflare/cfssl | CRYPTO | BSD-2-Clause | 31.7 MB | not yet graded | An agent building a PKI/CA service queries this for Cloudflare's CFSSL certificate signing, bundling and OCSP tooling. |
| chamber | segmentio/chamber | CRYPTO | MIT | 275.5 KB | not yet graded | An agent managing secrets via AWS SSM Parameter Store queries this for Chamber's parameter-store-backed CLI. |
| chaos-mesh | chaos-mesh/chaos-mesh | OPS | Apache-2.0 | 16.2 MB | not yet graded | An agent doing chaos engineering on Kubernetes queries this for Chaos Mesh fault-injection CRDs (pod/network/IO/time faults). |
| chaostoolkit-docs | chaostoolkit/chaostoolkit-documentation | OPS | Apache-2.0 | 1.3 MB | not yet graded | An agent designing chaos-engineering experiments queries this for the experiment format, probe/action API, and reliability-testing workflow. |
| chatdoctor | Kent0n-Li/ChatDoctor | HEALTH | Apache-2.0 | 26.8 MB | not yet graded | An agent building a patient-facing medical chat assistant queries this for 100k+ real doctor-patient dialogues (HealthCareMagic/iCliniq) and the fine-tuning rec |
| chdb | chdb-io/chdb | DB | Apache-2.0 | 17.8 MB | not yet graded | An agent working on embedding an OLAP engine in-process with no server queries this for chDB's approach to embedding ClickHouse as a library. |
| checkov | bridgecrewio/checkov | SEC | Apache-2.0 | 66.7 MB | not yet graded | an agent scanning infrastructure-as-code queries this for Checkov's Python/YAML policy checks across Terraform, CloudFormation and Kubernetes. |
| choosealicense-data | github/choosealicense.com | LAW | MIT | 1.1 MB | not yet graded | An agent building a license picker or classifying a project's license queries this for structured per-license metadata (permissions/conditions/limitations) and |
| chroma | chroma-core/chroma | DB | Apache-2.0 | 45.8 MB | not yet graded | DB |
| chronicle-detection-rules | chronicle/detection-rules | SEC | Apache-2.0 | 2.4 MB | not yet graded | an agent building Google SecOps content queries this for community YARA-L detection rules and dashboard templates. |
| cicero-contract-templates | accordproject/cicero-template-library | LAW | Apache-2.0 | 121.1 MB | not yet graded | law |
| circl | cloudflare/circl | CRYPTO | BSD-3-Clause | 51.9 MB | not yet graded | An agent working on modern or post-quantum crypto queries this for Cloudflare CIRCL's Go implementations of X25519/X448, hash-to-curve, BLS, Kyber/Dilithium and |
| cisa | xerj-org/corpus-cisa | SEC | US-government public domain (17 USC 105) and CC0-1.0 | 4.8 MB | G7 4.0/5 | One CISA corpus: advisories and the Known Exploited Vulnerabilities catalog together. An agent answering 'is this CVE exploited in the wild, and what do the mit |
| clipboard-apis | w3c/clipboard-apis | WEB | W3C-20150513 | 224.9 KB | not yet graded | An agent implementing copy/paste queries this for the Clipboard API's read/write model, formats and permission/transient-activation gating. |
| cloud-custodian | cloud-custodian/cloud-custodian | SEC | Apache-2.0 | 78.9 MB | not yet graded | an agent codifying cloud guardrails queries this for Cloud Custodian's YAML policy filters/actions across AWS, Azure and GCP. |
| cloudevents-spec | cloudevents/spec | API | Apache-2.0 | 32.7 MB | not yet graded | An agent working on event interchange between services queries this for the CloudEvents envelope format and its HTTP, Kafka, AMQP, and MQTT protocol bindings. |
| cloudnative-pg | cloudnative-pg/cloudnative-pg | OPS | Apache-2.0 | 59.1 MB | not yet graded | An agent running PostgreSQL on Kubernetes queries this for the operator's failover, backup, and rolling-update logic. |
| cluster-api | kubernetes-sigs/cluster-api | OPS | Apache-2.0 | 34.2 MB | not yet graded | An agent provisioning clusters declaratively queries this for the Cluster API machine/infrastructure provider model and lifecycle controllers. |
| cognita | truefoundry/cognita | AI | Apache-2.0 | 50.8 MB | not yet graded | An agent productionising modular RAG queries this for a config-driven framework with pluggable components. |
| commonaccord-legal-docs | CommonAccord/Cmacc-Org | LAW | MIT | 22.6 MB | not yet graded | law |
| commonpaper-contracts | CommonPaper/CSA | LAW | CC-BY-4.0 | 42.9 KB | not yet graded | law |
| complianceascode-content | ComplianceAsCode/content | LAW | BSD-3-Clause | 65.2 MB | not yet graded | An agent mapping security controls to regulations (STIG/CIS/800-53) queries this for machine-readable hardening content and regulatory crosswalks. |
| compose-spec | compose-spec/compose-spec | OPS | Apache-2.0 | 468.6 KB | not yet graded | An agent parsing or generating Compose files queries this for the authoritative Compose file schema and semantics. |
| concourse | concourse/concourse | OPS | Apache-2.0 | 37.1 MB | not yet graded | An agent building pipeline-based CI queries this for resource/job abstractions and reproducible build semantics. |
| congress-legislators | unitedstates/congress-legislators | LAW | CC0-1.0 | 12.3 MB | not yet graded | An agent resolving U.S. lawmakers, committees, or bill sponsors queries this for the CC0 dataset of current and historical members of Congress with cross-walk I |
| connect-go | connectrpc/connect-go | NET | Apache-2.0 | 1.8 MB | not yet graded | An agent building an RPC protocol queries this for the Connect/gRPC/gRPC-Web wire semantics and unary/streaming handler design. |
| connect-protocol | connectrpc/connectrpc.com | API | Apache-2.0 | 3.0 MB | not yet graded | An agent implementing a protobuf RPC protocol over HTTP/1.1 and HTTP/2 with gRPC compatibility queries this for the Connect protocol specification and docs. |
| containerd | containerd/containerd | OPS | Apache-2.0 | 55.3 MB | not yet graded | An agent operating a container runtime queries this for containerd's image/snapshot/task management and CRI implementation. |
| controlflow | PrefectHQ/ControlFlow | AI | Apache-2.0 | 2.6 MB | not yet graded | An agent embedding LLM steps inside deterministic workflows queries this for task-centric orchestration built on Prefect. |
| conventional-commits | conventional-commits/conventionalcommits.org | WRITE | MIT | 1.4 MB | not yet graded | An agent writing or validating commit messages queries this for the Conventional Commits specification (types, scopes, breaking-change wording) in many language |
| coredns | coredns/coredns | NET | Apache-2.0 | 4.0 MB | not yet graded | an agent working on a plugin-based DNS server queries this for the middleware chain and zone/record handling |
| coroot | coroot/coroot | OPS | Apache-2.0 | 71.3 MB | not yet graded | An agent building zero-instrumentation observability queries this for eBPF-based metric/trace capture and SLO/health inference. |
| counsel-chat | nbertagnolli/counsel-chat | HEALTH | MIT | 11.2 MB | not yet graded | An agent building mental-health counseling response generation queries this for licensed-therapist answers to client questions from the CounselChat platform. |
| courts-db | freelawproject/courts-db | LAW | BSD-2-Clause | 2.0 MB | not yet graded | An agent resolving or disambiguating U.S. court names queries this for the structured database of courts, jurisdictions, date ranges, and name variants. |
| covid-qa | deepset-ai/COVID-QA | HEALTH | Apache-2.0 | 17.6 MB | not yet graded | An agent building COVID-19 scientific-document QA queries this for ~2k SQuAD-style question-answer pairs annotated over CORD-19 articles. |
| cozo | cozodb/cozo | DB | MPL-2.0 | 5.5 MB | not yet graded | an agent working on an embeddable multi-model store queries this for relational-plus-graph-plus-vector (Datalog) query evaluation in Rust |
| cri-o | cri-o/cri-o | OPS | Apache-2.0 | 100.8 MB | not yet graded | An agent running a Kubernetes-dedicated container runtime queries this for CRI-O's CRI implementation and OCI integration. |
| crossplane | crossplane/crossplane | OPS | Apache-2.0 | 21.7 MB | not yet graded | An agent building control-plane IaC queries this for Crossplane compositions, provider resources and reconciliation. |
| csp | w3c/webappsec-csp | WEB | W3C-20150513 | 3.9 MB | not yet graded | An agent hardening a web surface queries this for Content Security Policy directive grammar, source-list matching and nonce/hash enforcement. |
| css-tree | csstree/csstree | WEB | MIT | 2.0 MB | not yet graded | An agent parsing/validating CSS queries this for csstree's fast CSS tokenizer, AST, selector parsing and value-syntax matching. |
| ctakes | apache/ctakes | HEALTH | Apache-2.0 | 43.2 MB | not yet graded | An agent extracting concepts from clinical notes queries this for Apache cTAKES pipelines: section detection, negation/context, and UMLS dictionary lookup. |
| cue | cue-lang/cue | OPS | Apache-2.0 | 15.9 MB | not yet graded | An agent validating or generating configuration queries this for CUE's type/constraint model and schema unification. |
| curl-src | curl/curl | NET | curl | 17.6 MB | not yet graded | an agent writing an HTTP client queries this for protocol handling, auth negotiation and the protocol-selection seams |
| curve25519-dalek | dalek-cryptography/curve25519-dalek | CRYPTO | BSD-3-Clause | 2.6 MB | not yet graded | An agent working on elliptic-curve or signature crypto in Rust queries this for the dalek curve25519/ed25519/x25519 field-arithmetic and group-operation code. |
| cyclonedx-spec | CycloneDX/specification | SEC | Apache-2.0 | 7.5 MB | not yet graded | an agent producing SBOMs queries this for the CycloneDX JSON/XML schema and field semantics. |
| dagger | dagger/dagger | OPS | Apache-2.0 | 173.1 MB | not yet graded | An agent writing portable CI pipelines queries this for Dagger's programmable container-based pipeline engine and caching. |
| dailymed | xerj-org/corpus-dailymed | HEALTH | US-government public domain (NLM), attribution requested | 26.8 MB | G7 5.0/5 | an agent answering medication questions queries this for what the FDA-approved label actually says: indications, do-not-crush/dose-form rules, epidural-timing w |
| datafusion | apache/datafusion | DB | Apache-2.0 | 58.3 MB | not yet graded | DB |
| datasketches-cpp | apache/datasketches-cpp | DB | Apache-2.0 | 2.7 MB | not yet graded | An agent working on approximate aggregations (distinct-count, quantiles, frequent-items) queries this for DataSketches' HLL, CPC, KLL and Theta sketch algorithm |
| deepeval | confident-ai/deepeval | AI | Apache-2.0 | 42.1 MB | not yet graded | An agent unit-testing LLM outputs queries this for pytest-style LLM metrics and assertions. |
| delta-rs | delta-io/delta-rs | DB | Apache-2.0 | 15.4 MB | not yet graded | An agent working on a transactional table format (ACID over object storage) queries this for delta-rs's transaction log, snapshot isolation and parquet-file man |
| density | g1mv/density | NET | Apache-2.0 | 9.8 MB | not yet graded | An agent implementing a very fast byte-oriented compressor queries this for the Chameleon/Cheetah/Lion kernels and streaming API design. |
| dgraph | hypermodeinc/dgraph | DB | Apache-2.0 | 65.4 MB | not yet graded | An agent working on a distributed graph database with a sharded posting-list store queries this for Dgraph's predicate-sharded storage and query execution in Go |
| digitalocean-openapi | digitalocean/openapi | API | Apache-2.0 | 3.8 MB | not yet graded | An agent modelling a cloud-provider control-plane REST API queries this for DigitalOcean's OpenAPI structure, $refs and bundling setup. |
| diskann | microsoft/DiskANN | DB | MIT | 12.4 MB | not yet graded | An agent working on disk-resident billion-scale ANN queries this for DiskANN's Vamana graph build, PQ compression and SSD-aware search. |
| dispatch | Netflix/dispatch | OPS | Apache-2.0 | 13.2 MB | not yet graded | An agent building incident-response tooling queries this for Netflix Dispatch's incident workflow, roles and integrations. |
| distilabel | argilla-io/distilabel | AI | Apache-2.0 | 10.0 MB | not yet graded | An agent building synthetic training data queries this for pipeline-based data generation and AI-feedback labeling. |
| dnscrypt-proxy | DNSCrypt/dnscrypt-proxy | NET | ISC | 15.0 MB | not yet graded | an agent working on encrypted DNS (DNSCrypt/DoH) queries this for protocol framing and resolver forwarding |
| docassemble | jhpyle/docassemble | LAW | MIT | 72.8 MB | not yet graded | An agent building guided legal interviews and automated document assembly queries this for a mature open-source document-automation platform. |
| docker-docs | docker/docs | OPS | Apache-2.0 | 48.5 MB | not yet graded | An agent writing Dockerfiles and compose files queries this for the canonical Docker CLI, build and Compose reference docs. |
| dragonboat | lni/dragonboat | DB | Apache-2.0 | 2.5 MB | not yet graded | an agent implementing multi-raft queries this for the sharded log and snapshot plumbing |
| drpc | storj/drpc | NET | MIT | 809.2 KB | not yet graded | An agent designing a lightweight RPC transport queries this for a minimal stream-multiplexed protocol and codegen that avoids HTTP/2. |
| dspy | stanfordnlp/dspy | AI | MIT | 25.1 MB | not yet graded | An agent building a prompt-optimization or declarative LLM-program pipeline queries this for DSPy's signatures, modules, and optimizers (compile-prompts-from-me |
| duckdb | duckdb/duckdb | DB | MIT | 245.6 MB | not yet graded | an agent working on query execution queries this for how a real embeddable OLAP engine does join ordering, expression rewrites and memory-aware operators |
| dunorth595--ai-agent-chat | DuNorth595/ai_agent_chat | AI | MIT | 827.1 KB | not yet graded | an agent building a voice interface a user can interrupt queries this for how DuNorth595/ai_agent_chat implements full-duplex voice with barge-in -- the TTS is |
| dunorth595--ai-agent-memory | DuNorth595/ai_agent_memory | AI | MIT | 383.2 KB | not yet graded | an agent needing memory that survives between sessions queries this for how DuNorth595/ai_agent_memory implements a portable long-term memory as one index file |
| dunorth595--xrpl-agent-id | DuNorth595/xrpl_agent_id | AI | MIT | 1.1 MB | not yet graded | an agent giving an AI agent a verifiable on-chain identity queries this for how DuNorth595/xrpl_agent_id implements did:xrpl identifiers, W3C DID Documents unde |
| dunorth595--xrpl-agent-log | DuNorth595/xrpl_agent_log | OPS | MIT | 316.8 KB | not yet graded | an agent needing a tamper-evident history for AI agent actions queries this for how DuNorth595/xrpl_agent_log implements a hash-chained, mutually-signed append- |
| dunorth595--xrpl-did-mcp | DuNorth595/xrpl_did_mcp | AI | MIT | 46.8 KB | not yet graded | an MCP-capable agent needing did:xrpl operations queries this for how DuNorth595/xrpl_did_mcp exposes DID derivation and DID Document publication as MCP tools, |
| dwyl-english-words | dwyl/english-words | WRITE | Unlicense | 18.5 MB | not yet graded | An agent needing a plain English word corpus (validation, vocabulary, autocomplete) queries this for ~470k English words as text/JSON. |
| ecfr-title-12 | xerj-org/corpus-ecfr-title-12 | LAW | LicenseRef-PublicDomain-USGov | 24.4 MB | G7 5.0/5 | an agent answering consumer-credit or mortgage questions queries this for what Regulation Z / TILA actually says: higher-priced APR triggers, ability-to-repay/Q |
| ecfr-title-14 | xerj-org/corpus-ecfr-title-14 | LAW | US-government public domain (CFR text, 17 USC 105) | 11.2 MB | G7 5.0/5 | a pilot or mechanic queries this for the current FAR text: visibility minimums, transponder airspace, inspection intervals, right-of-way |
| ecfr-title-26 | xerj-org/corpus-ecfr-title-26 | LAW | US-government-public-domain | 70.7 MB | G7 1.0/5 | an agent answering federal tax questions queries this for what the Treasury regulations actually say: de minimis safe harbor, estimated-tax safe harbor, home-of |
| ecfr-title-29 | xerj-org/corpus-ecfr-title-29 | LAW | LicenseRef-PublicDomain-USGov | 23.1 MB | G7 5.0/5 | an agent answering workplace/employment questions queries this for what 29 CFR actually says: OSHA guardrail height, FMLA 12-month/1250-hour eligibility, FLSA o |
| ecma262 | xerj-org/corpus-ecma262-sections | WEB | Ecma alternative notice (spec text; W3C-derived) + BSD-style (code) | 3.4 MB | G7 3.0/5 | an agent building a JS-adjacent tool queries this for the spec's abstract operations and completion-record semantics |
| edliz-zimbabwe-treatment-guidelines | michaeldera/edliz | HEALTH | MIT | 9.7 MB | not yet graded | health |
| ell | MadcowD/ell | AI | MIT | 24.1 MB | not yet graded | An agent treating prompts as versioned functions queries this for the language-model-programs model with autosave/versioning and visualization. |
| emp-tool | emp-toolkit/emp-tool | CRYPTO | Apache-2.0 | 12.2 MB | not yet graded | An agent building secure two-party/MPC computation queries this for the EMP-toolkit garbled-circuit and oblivious-transfer primitives. |
| enet | lsalzman/enet | NET | MIT | 369.2 KB | not yet graded | an agent working on reliable UDP for games queries this for packet sequencing, channels, and connection management |
| envoy | envoyproxy/envoy | NET | Apache-2.0 | 161.4 MB | not yet graded | an agent building a data plane queries this for filter-chain, xDS and overload-manager internals |
| errata-ai-readability-vale | errata-ai/Readability | WRITE | MIT | 18.5 KB | not yet graded | An agent building prose-linting or readability features queries this Vale style for how to encode Flesch-Kincaid/Gunning-Fog/SMOG metrics as lint rules. |
| esbuild | evanw/esbuild | WEB | MIT | 9.0 MB | not yet graded | An agent building a fast JS/TS bundler queries this for esbuild's parser, scope/linker design and tree-shaking - a reference for speed-oriented bundling. |
| eset-malware-ioc | eset/malware-ioc | SEC | BSD-2-Clause | 12.8 MB | not yet graded | An agent operationalizing threat-research output queries this for ESET investigation IOCs: YARA (.yar), Snort (.rules) and sample-hash lists tied to named campa |
| etcd-docs | etcd-io/website | OPS | CC-BY-4.0 | 32.8 MB | G7 5.0/5 | an agent operating a distributed system queries this for etcd's failure/restore/compaction semantics and tuning guides |
| etcd-src | etcd-io/etcd | DB | Apache-2.0 | 14.5 MB | not yet graded | an agent working on coordination stores queries this for raft integration, watch semantics and backend compaction |
| etherparse | JulianSchmid/etherparse | NET | Apache-2.0 | 3.3 MB | not yet graded | an agent working on packet classification queries this for zero-copy Ethernet/IPv4/IPv6/TCP/UDP header parsing and construction |
| eurlex-core | xerj-org/corpus-eurlex-core | LAW | NONE-FOUND | 1.9 MB | G7 4.0/5 | GDPR / AI Act / DSA / DMA / NIS2 article and recital text |
| exploit-pocs-2023 | 0-d3y/CVE-2023-29489, 04Shivam/CVE-2023-30253-Exploit, 0SPwn/CVE-2023-27372-PoC, 0axz-tools/CVE-2023-28121 | SEC | AGPL+Apache-2.0+Apache-2.0/BSD+BSD+GPL+GPL/MIT+LGPL+MIT+MPL-2.0+NONE-FOUND+SSPL+UNKNOWN | 33.0 GB | not yet graded | year backfill slice of the exploit-PoC set: 2023 PoC repositories at pinned shas |
| exploit-pocs-2024 | 007CRIPTOGRAFIA/c-CVE-2024-0044, 00xCanelo/CVE-2024-47533-PoC, 0Nightsedge0/CVE-2024-9950-PoC, 0XFFFF-XD/CVE-2024-4577-PHP-CGI-RCE | SEC | AGPL+Apache-2.0+BSD+GPL+LGPL+MIT+NONE-FOUND+UNKNOWN | 7.6 GB | not yet graded | year backfill slice of the exploit-PoC set: 2024 PoC repositories at pinned shas |
| exploit-pocs-2025 | 0-d3y/langflow-rce-exploit, 00lucasm/CVE-2025-58434-Flowiseai-Auth-Bypass-PoC, 00xCanelo/CVE-2025-27591, 00xCanelo/CVE-2025-32778 | SEC | AGPL+Apache-2.0+BSD+GPL+GPL/MIT+LGPL+MIT+MPL-2.0+NONE-FOUND+UNKNOWN | 9.8 GB | not yet graded | year backfill slice of the exploit-PoC set: 2025 PoC repositories at pinned shas |
| exploit-pocs-2026 | gabrielftanaka/CVE-2026-82221-PoC, rafabd1/CVE-2026-75604-poc, HackSpeak/CVE-2026-75604, e4zyy/Project-CVE-2026-75604 | SEC | AGPL+Apache-2.0+Apache-2.0/LGPL+Apache-2.0/MIT+BSD+GPL+LGPL+LGPL/MIT+MIT+MPL-2.0+NONE-FOUND+UNKNOWN | 4.3 GB | not yet graded | an agent triaging a fresh CVE queries this for runnable 2026 PoC repositories: exploit source, affected product, and the fix commit together |
| exploitdb-src | SEC | GPL-2.0 | 174.4 MB | not yet graded | an agent matching an exploit to its target queries this for the full Exploit-DB record and exploit source | |
| external-secrets | external-secrets/external-secrets | CRYPTO | Apache-2.0 | 36.5 MB | not yet graded | An agent syncing secrets from cloud vaults into Kubernetes queries this for the External Secrets Operator's provider abstraction. |
| eyecite | freelawproject/eyecite | LAW | BSD-2-Clause | 679.6 KB | not yet graded | An agent extracting legal citations from judicial text queries this for a battle-tested citation-recognition/resolution implementation run across millions of op |
| faah | kiron0/faah | WEB | MIT | 1.6 MB | not yet graded | an agent building VS Code developer-feedback extensions queries this for terminal execution monitoring, editor diagnostics, audio alerts, settings webviews and |
| fabric | danielmiessler/fabric | AI | MIT | 27.5 MB | not yet graded | An agent building a reusable prompt library queries this for battle-tested, task-specific prompt patterns and their markdown structure. |
| faiss | facebookresearch/faiss | DB | MIT | 9.9 MB | not yet graded | an agent working on ANN queries this for index construction, IVF/HNSW composition and GPU paths in the reference implementation |
| falco-rules | falcosecurity/rules | SEC | Apache-2.0 | 730.2 KB | not yet graded | security |
| fastchat | lm-sys/FastChat | AI | Apache-2.0 | 15.1 MB | not yet graded | An agent serving and judging chat models queries this for serving, MT-Bench, and LLM-as-judge/Arena patterns. |
| fastcrypto | MystenLabs/fastcrypto | CRYPTO | Apache-2.0 | 3.5 MB | not yet graded | an agent working on a performance-critical crypto layer for a blockchain/database queries this for Mysten's vetted wrappers over signatures, hashing and VRFs |
| fastmcp | jlowin/fastmcp | AI | Apache-2.0 | 58.7 MB | not yet graded | An agent standing up an MCP server in Python queries this for a high-level MCP server/client framework. |
| fastrag | IntelLabs/fastRAG | AI | Apache-2.0 | 27.1 MB | not yet graded | An agent optimising RAG for efficiency queries this for retrieval acceleration and context-compression techniques. |
| fda-food-code | xerj-org/corpus-fda-food-code | HEALTH | US-government public domain (FDA/HHS, 17 USC 105) | 6.3 MB | G7 4.0/5 | FDA Food Code 2022 provisions: cooking/holding temperatures, sanitizer tables, reporting rules |
| fhir-test-cases | FHIR/fhir-test-cases | HEALTH | Apache-2.0 | 191.9 MB | not yet graded | An agent implementing a FHIR server or validator queries this for canonical example resources and conformance test cases across FHIR R3/R4/R5. |
| flagembedding | FlagOpen/FlagEmbedding | AI | MIT | 55.0 MB | not yet graded | An agent choosing or fine-tuning retrieval models queries this for BGE embedding and reranker training/usage. |
| flatbuffers | google/flatbuffers | NET | Apache-2.0 | 13.0 MB | not yet graded | an agent designing zero-copy serialisation queries this for the vtable layout and verifier approach |
| flexrag | ictnlp/FlexRAG | AI | MIT | 17.7 MB | not yet graded | An agent reproducing/evaluating RAG systems queries this for a flexible research framework spanning text, multimodal, and web-accessible RAG. |
| fluent-bit | fluent/fluent-bit | OPS | Apache-2.0 | 165.4 MB | not yet graded | An agent collecting logs and metrics at the edge queries this for Fluent Bit's lightweight C pipeline, parsers and routing. |
| fluentd | fluent/fluentd | OPS | Apache-2.0 | 4.1 MB | not yet graded | An agent building log pipelines queries this for Fluentd input/filter/output plugin routing and buffering. |
| flux2 | fluxcd/flux2 | OPS | Apache-2.0 | 3.1 MB | not yet graded | An agent building pull-based GitOps queries this for Flux source/kustomize/helm controllers and reconciliation semantics. |
| fluxcd-website | fluxcd/website | OPS | Apache-2.0 | 111.5 MB | not yet graded | An agent adopting Flux queries this for GitOps setup guides, controller reference and migration docs. |
| foodon | FoodOntology/foodon | HEALTH | CC-BY-4.0 | 247.7 MB | not yet graded | An agent working on normalizing food and ingredient names queries this for a controlled vocabulary of foods, food products, and source/processing relationships |
| fortio | fortio/fortio | OPS | Apache-2.0 | 939.5 KB | not yet graded | An agent benchmarking HTTP/gRPC services queries this for load generation, latency histograms, and percentile reporting. |
| fory | apache/fory | NET | Apache-2.0 | 37.1 MB | not yet graded | an agent working on fast cross-language serialization queries this for reference tracking, schema evolution, and JIT-accelerated codecs |
| freebsd-handbook | freebsd/freebsd-doc | OPS | FreeBSD-Doc | 361.8 MB | not yet graded | an agent administering a BSD system queries this for the handbook's exact rc/ZFS/jail/networking procedures |
| garak | NVIDIA/garak | AI | Apache-2.0 | 11.2 MB | not yet graded | An agent red-teaming an LLM queries this for probe/detector implementations for jailbreaks and prompt injection. |
| gatekeeper-library | open-policy-agent/gatekeeper-library | SEC | Apache-2.0 | 3.6 MB | not yet graded | an agent writing Kubernetes admission policy queries this for OPA Gatekeeper constraint templates (Rego) and their test cases. |
| gateway-api | kubernetes-sigs/gateway-api | OPS | Apache-2.0 | 17.8 MB | not yet graded | An agent implementing ingress/gateway routing queries this for the Gateway API spec, conformance semantics, and resource model. |
| gatling | gatling/gatling | OPS | Apache-2.0 | 4.4 MB | not yet graded | An agent building performance test suites queries this for Gatling's DSL scenarios, injection profiles and assertions. |
| gdpr-commonaccord | CommonAccord/GDPR | LAW | MIT | 907.5 KB | not yet graded | law |
| gemini-cookbook | google-gemini/cookbook | AI | Apache-2.0 | 142.8 MB | not yet graded | An agent building on Gemini queries this for recipes on long context, multimodal input, and structured output. |
| getrandom | rust-random/getrandom | CRYPTO | Apache-2.0 | 214.7 KB | not yet graded | An agent sourcing OS randomness portably queries this for getrandom's per-platform syscall backends. |
| ghsa-db | github/advisory-database | SEC | CC-BY-4.0 | 75.7 MB | G7 5.0/5 | an agent auditing dependencies across ecosystems queries this for GitHub-reviewed advisories with CVSS and patched ranges |
| giskard | Giskard-AI/giskard | AI | Apache-2.0 | 10.4 MB | not yet graded | An agent scanning an LLM app for vulnerabilities queries this for automated test/red-team scan implementations. |
| github-api-description | github/rest-api-description | API | MIT | 7.3 GB | G7 3.0/5 | an agent building a GitHub integration queries this for the machine-readable schema of a huge production REST API (the exemplar OpenGraph/OpenAPI corpus) |
| github-opensource-guide | github/opensource.guide | WRITE | CC-BY-4.0 | 14.2 MB | not yet graded | An agent writing open-source project docs, CONTRIBUTING guides and maintainer communications queries this for editorial best practices and templates. |
| github-site-policy | github/site-policy | LAW | CC0-1.0 | 649.6 KB | not yet graded | An agent drafting SaaS terms of service, acceptable-use, or privacy text queries this for GitHub's full suite of legal site policies in Markdown with version hi |
| gitlab-runbooks | OPS | MIT | 155.4 MB | G7 3.0/5 | an agent on call for a SaaS queries this for real alert definitions, escalations and mitigation playbooks from a large production fleet | |
| glasskube | glasskube/glasskube | OPS | Apache-2.0 | 1.7 MB | not yet graded | An agent building a Kubernetes package manager queries this for dependency-aware install/update and GitOps package distribution. |
| gnark | Consensys/gnark | CRYPTO | Apache-2.0 | 17.6 MB | not yet graded | An agent building ZK circuits in Go queries this for gnark's R1CS/PLONK frontends and backends. |
| gnostic | google/gnostic | API | Apache-2.0 | 2.6 MB | not yet graded | An agent converting OpenAPI/Discovery docs into protobuf models queries this for gnostic's spec parsing and plugin pipeline. |
| go-jose | go-jose/go-jose | CRYPTO | Apache-2.0 | 882.1 KB | not yet graded | An agent implementing JWS/JWE/JWK queries this for go-jose's JOSE object serialization and algorithm support. |
| go-stomp | go-stomp/stomp | NET | Apache-2.0 | 222.0 KB | not yet graded | an agent working on a STOMP messaging client queries this for the text frame grammar and subscription/ack semantics |
| go-tpm | google/go-tpm | CRYPTO | Apache-2.0 | 1.5 MB | not yet graded | An agent driving a TPM from Go queries this for go-tpm's TPM 1.2/2.0 command marshalling and session handling. |
| go-vulndb | golang/vulndb | SEC | BSD-3-Clause | 4.1 MB | G7 4.0/5 | an agent reviewing Go modules queries this for Go-team-curated advisories with symbols and fixed versions |
| golang-jwt | golang-jwt/jwt | CRYPTO | MIT | 235.3 KB | not yet graded | An agent issuing/validating JWTs in Go queries this for golang-jwt's signing methods and claims validation. |
| golang-x-crypto | golang/crypto | CRYPTO | BSD-3-Clause | 5.2 MB | not yet graded | an agent working on crypto primitives in Go queries this for x/crypto's argon2/scrypt/chacha20poly1305/ssh/ocsp/acme implementations |
| good-docs-templates | thegooddocsproject/templates | WRITE | 0BSD | 22.7 MB | not yet graded | an agent structuring a document queries this for how-to, tutorial, reference and explanation templates with embedded writing guidance |
| google-aip | aip-dev/google.aip.dev | API | CC-BY-4.0 | 806.4 KB | not yet graded | An agent designing a resource-oriented REST or gRPC API queries this for Google's API Improvement Proposals on naming, pagination, field masks, errors, and long |
| google-devdocs-vale | vale-cli/Google | WRITE | MIT | 54.6 KB | not yet graded | an agent writing developer docs queries this for Google dev-doc style rules: voice, tense, second person, word choice |
| google-eng-practices | google/eng-practices | API | CC-BY-3.0 | 89.5 KB | G7 4.0/5 | an agent reviewing code queries this for the review standard's concrete what-to-look-for and clangers |
| google-fuzzing | google/fuzzing | SEC | Apache-2.0 | 1.6 MB | not yet graded | an agent setting up fuzzing queries this for fuzzer dictionaries, tutorials and best-practice documentation (libFuzzer/AFL). |
| google-licenseclassifier | google/licenseclassifier | LAW | Apache-2.0 | 4.3 MB | not yet graded | An agent building fuzzy license identification queries this for a Go classifier plus its embedded corpus of known license and header texts. |
| googleapis | googleapis/googleapis | API | Apache-2.0 | 76.7 MB | not yet graded | An agent designing resource-oriented gRPC/REST APIs the Google way queries this for canonical service .proto definitions and the common google.api protos. |
| gopass | gopasspw/gopass | CRYPTO | MIT | 4.9 MB | not yet graded | An agent building a team password store queries this for gopass' pass-compatible, GPG/age-backed secret store. |
| goreleaser | goreleaser/goreleaser | OPS | MIT | 6.7 MB | not yet graded | An agent automating release builds queries this for cross-compilation, packaging, signing and changelog/release publishing config. |
| gorilla | ShishirPatil/gorilla | AI | Apache-2.0 | 178.9 MB | not yet graded | An agent designing function/tool calling queries this for API-invocation training and the function-calling leaderboard. |
| gosec-rules | securego/gosec | SEC | Apache-2.0 | 1.8 MB | not yet graded | An agent writing Go SAST or secure-coding checks queries this for gosec's rule implementations (AST/SSA patterns for SQLi, command injection, weak RNG, TLS misc |
| goss | goss-org/goss | OPS | Apache-2.0 | 774.1 KB | not yet graded | An agent validating server and infrastructure state queries this for Goss/gossfile declarative health and readiness checks. |
| got | sindresorhus/got | NET | MIT | 2.5 MB | not yet graded | An agent implementing an HTTP client with retries and redirects queries this for retry/backoff policy, redirect handling, and streaming response semantics. |
| govuk-design-system | xerj-org/corpus-govuk-components | WEB | MIT | 392.6 KB | G7 0.0/5 | an agent building accessible UI queries this for the component's exact accessibility criteria and edge cases |
| gpt-engineer | gpt-engineer-org/gpt-engineer | AI | MIT | 16.0 MB | not yet graded | An agent scaffolding a project from a spec queries this for spec-to-codebase generation loops. |
| grafana-alloy | grafana/alloy | OPS | Apache-2.0 | 38.3 MB | not yet graded | An agent deploying a vendor-neutral telemetry collector queries this for OTel/Prometheus pipeline components and configuration. |
| graphql-composite-schemas-spec | graphql/composite-schemas-spec | API | MIT | 380.4 KB | not yet graded | An agent federating multiple GraphQL services into one supergraph queries this for the Composite Schemas spec (keys, lookups, schema composition). |
| graphql-over-http | graphql/graphql-over-http | API | MIT | 602.2 KB | not yet graded | An agent serving GraphQL over HTTP queries this for request/response media types, status-code and transport conventions. |
| graphql-spec | graphql/graphql-spec | API | JDF-charter (spec text (c) Facebook 2015-2018 + JDF) | 5.0 MB | G7 5.0/5 | an agent building a GraphQL server queries this for execution, defer/stream and introspection semantics |
| graphrag | microsoft/graphrag | AI | MIT | 17.0 MB | not yet graded | An agent doing knowledge-graph-grounded RAG queries this for entity/community extraction and graph-based retrieval. |
| griptape | griptape-ai/griptape | AI | Apache-2.0 | 20.3 MB | not yet graded | An agent building modular pipelines queries this for structures (agents/pipelines/workflows), tasks, and tool/driver abstractions. |
| grpc-gateway | grpc-ecosystem/grpc-gateway | API | BSD-3-Clause | 9.4 MB | not yet graded | An agent exposing gRPC services as REST/JSON queries this for the google.api.http annotation mapping and reverse-proxy generation. |
| grpc-go | grpc/grpc-go | NET | Apache-2.0 | 14.1 MB | not yet graded | NET |
| grpc-proposals | grpc/proposal | API | Apache-2.0 | 10.0 MB | not yet graded | An agent working on gRPC protocol or framework behaviour queries this for the gRFC design proposals that define wire format, naming, retries, and cross-language |
| guardrails | guardrails-ai/guardrails | AI | Apache-2.0 | 15.0 MB | not yet graded | An agent validating and correcting LLM output queries this for RAIL schemas and output validators/reask loops. |
| guidance | guidance-ai/guidance | AI | MIT | 36.5 MB | not yet graded | An agent constraining LLM output to a grammar, template, or control flow queries this for guidance's constrained-generation primitives and token-healing. |
| h2 | hyperium/h2 | NET | MIT | 58.5 MB | not yet graded | an agent working on an HTTP/2 protocol core queries this for stream multiplexing, flow control, and HPACK framing |
| h2o | h2o/h2o | NET | MIT | 30.2 MB | not yet graded | An agent implementing a high-performance HTTP/1-2 server queries this for event-loop request handling, HTTP/2 prioritization, and header processing in C. |
| hacl-star | hacl-star/hacl-star | CRYPTO | Apache-2.0 | 90.9 MB | not yet graded | An agent needing formally-verified crypto queries this for HACL*/EverCrypt's verified C implementations and agile API. |
| harbor | goharbor/harbor | OPS | Apache-2.0 | 18.8 MB | not yet graded | An agent running a private registry queries this for Harbor image storage, replication, scanning and RBAC. |
| harper | Automattic/harper | WRITE | Apache-2.0 | 47.5 MB | not yet graded | An agent building a fast local grammar/style checker in Rust queries this for its rule engine, lint rules and curated dictionary (no cloud, no LLM). |
| haystack | deepset-ai/haystack | AI | Apache-2.0 | 101.3 MB | not yet graded | An agent building a production RAG/NLP pipeline queries this for composable retriever/reader/generator components. |
| head-qa | aghie/head-qa | HEALTH | MIT | 69.0 KB | not yet graded | An agent building healthcare professional-exam QA queries this for HEAD-QA: Spanish Ministry of Health specialization-exam questions (medicine, nursing, pharmac |
| helm | helm/helm | OPS | Apache-2.0 | 4.0 MB | not yet graded | An agent packaging Kubernetes apps queries this for chart templating, release lifecycle and rollback, and dependency/values resolution. |
| hivemq-mqtt-client | hivemq/hivemq-mqtt-client | NET | Apache-2.0 | 3.4 MB | not yet graded | NET |
| how-they-sre | upgundecha/howtheysre | OPS | CC0-1.0 | 305.5 KB | not yet graded | An agent studying real-world SRE queries this for curated summaries of how many named organizations run reliability engineering. |
| httparse | seanmonstar/httparse | NET | Apache-2.0 | 272.5 KB | not yet graded | an agent working on a zero-copy HTTP/1.x request/response parser queries this for push-parser state machines and header handling |
| httpx | encode/httpx | NET | BSD-3-Clause | 3.2 MB | not yet graded | An agent designing an HTTP client queries this for sync/async client design, connection pooling, and HTTP/2 transport behavior. |
| huggingface-hub-docs | huggingface/hub-docs | AI | Apache-2.0 | 20.3 MB | not yet graded | An agent writing model/dataset cards queries this for the Hugging Face model-card spec and metadata conventions. |
| hyper | hyperium/hyper | NET | MIT | 1.4 MB | not yet graded | An agent building an async HTTP/1 and HTTP/2 stack in Rust queries this for connection handling, request/response encoding, and the client/server state machines |
| ibm-openapi-validator | IBM/openapi-validator | API | Apache-2.0 | 2.9 MB | not yet graded | An agent validating OpenAPI documents against configurable governance rules queries this for IBM's validator rule set and scoring model. |
| icd10-who | StefanoTrv/simple_icd_10 | HEALTH | CC0-1.0 | 6.9 MB | not yet graded | health |
| icd10cm-codes | bryand1/icd10-cm | HEALTH | MIT | 731.8 KB | not yet graded | health |
| icd9-icd10-hierarchy | LuChang-CS/icd_hierarchical_structure | HEALTH | MIT | 190.0 MB | not yet graded | health |
| iceberg | apache/iceberg | DB | Apache-2.0 | 59.5 MB | not yet graded | an agent building lakehouse tables queries this for spec semantics and the reference metadata/manifest handling |
| in-solidarity-bot | jpoehnelt/in-solidarity-bot | WRITE | Apache-2.0 | 806.9 KB | not yet graded | An agent enforcing inclusive language in a repo queries this for its default term->replacement rules and configuration model. |
| influxdb | influxdata/influxdb | DB | Apache-2.0 OR MIT | 26.1 MB | not yet graded | an agent working on time series queries this for TSM/flux paths and downsampling semantics |
| infracost | infracost/infracost | OPS | Apache-2.0 | 15.7 MB | not yet graded | An agent estimating cloud cost in CI queries this for Infracost's Terraform cost-diff pricing and policy checks. |
| inspect-ai | UKGovernmentBEIS/inspect_ai | AI | MIT | 51.4 MB | not yet graded | An agent building rigorous model evaluations queries this for a task/solver/scorer evaluation framework. |
| instructor | 567-labs/instructor | AI | MIT | 73.8 MB | not yet graded | An agent extracting typed/structured data from LLM calls queries this for instructor's Pydantic response-model patterns, retries, and validation hooks. |
| ipsum-blocklist | stamparm/ipsum | SEC | Unlicense | 4.5 MB | not yet graded | An agent building IP reputation/blocklist features queries this for a daily-aggregated list of malicious IPs ranked by number of source blocklists. |
| jaeger-docs | jaegertracing/documentation | OPS | Apache-2.0 | 32.8 MB | not yet graded | An agent instrumenting or operating distributed tracing queries this for Jaeger deployment, sampling configuration, and trace analysis. |
| joblint | rowanmanning/joblint | WRITE | MIT | 405.2 KB | not yet graded | An agent writing or auditing job descriptions queries this for rules that flag sexist, elitist or exclusionary wording with suggested replacements. |
| jose-js | panva/jose | CRYPTO | MIT | 1.6 MB | not yet graded | An agent implementing JOSE in JS/TS queries this for panva/jose's JWS/JWE/JWK handling and runtime-agnostic crypto. |
| jsdom | jsdom/jsdom | WEB | MIT | 44.9 MB | not yet graded | An agent implementing DOM/HTML semantics outside a browser queries this for a JS implementation of DOM, HTML elements, events and the WHATWG algorithms. |
| json-api-spec | json-api/json-api | API | CC0-1.0 | 2.0 MB | not yet graded | An agent building a JSON REST API queries this for the JSON:API specification's document structure, resource relationships, includes, and query-parameter conven |
| json-schema-spec | json-schema-org/json-schema-spec | API | BSD-2-Clause | 742.2 KB | G7 3.5/5 | an agent writing validators queries this for the dialect's exact keyword semantics (2019-09 vs 2020-12 differences bite) |
| json-schema-test-suite | json-schema-org/JSON-Schema-Test-Suite | API | MIT | 3.4 MB | not yet graded | An agent implementing a JSON Schema validator queries this for the canonical cross-draft conformance test cases (draft-04 through 2020-12). |
| jsonformer | 1rgs/jsonformer | AI | MIT | 300.4 KB | not yet graded | An agent forcing valid JSON from a local model queries this for token-level structured decoding over a schema. |
| jsonwebtoken-rust | Keats/jsonwebtoken | CRYPTO | MIT | 263.9 KB | not yet graded | An agent issuing/validating JWTs in Rust queries this for Keats/jsonwebtoken's encode/decode and algorithm handling. |
| jwx | lestrrat-go/jwx | CRYPTO | MIT | 3.3 MB | not yet graded | an agent working on a complete JWx stack in Go queries this for JWK set handling, JWS/JWE and JWT validation flows |
| kafka-protocol | xerj-org/corpus-kafka-docs | DB | Apache-2.0 | 3.0 MB | G7 3.0/5 | an agent implementing a Kafka client queries this for the protocol guide's exact request/response schemas and error codes |
| kafka-src | apache/kafka | DB | Apache-2.0 | 83.3 MB | not yet graded | an agent working on log systems queries this for the log-cleaner, controller and replication protocol internals |
| karma | prymitive/karma | OPS | Apache-2.0 | 24.2 MB | not yet graded | An agent building an alert dashboard queries this for Alertmanager aggregation, grouping, and silence-management UI patterns. |
| karpenter | kubernetes-sigs/karpenter | OPS | Apache-2.0 | 17.7 MB | not yet graded | An agent tuning Kubernetes node autoscaling and bin-packing queries this for NodePool/provisioner logic, consolidation, and instance-selection heuristics. |
| keda | kedacore/keda | OPS | Apache-2.0 | 9.0 MB | not yet graded | An agent configuring event-driven autoscaling queries this for KEDA scalers, ScaledObjects and the HPA/external-metrics bridge. |
| keep-a-changelog | olivierlacan/keep-a-changelog | WRITE | MIT | 5.2 MB | not yet graded | An agent writing or standardizing CHANGELOG files queries this for the Keep a Changelog convention (sections, versioning, wording). |
| keywhiz | square/keywhiz | CRYPTO | Apache-2.0 | 2.4 MB | not yet graded | An agent designing a centralized secrets-distribution service queries this for Square's Keywhiz server and FUSE-based secret delivery. |
| kics | Checkmarx/kics | SEC | Apache-2.0 | 122.9 MB | not yet graded | an agent finding IaC misconfigurations queries this for KICS query definitions (Rego-style) plus their vulnerable/expected test fixtures. |
| kkp-asymptotic-spectral-bound | shaikidris/kkp-asymptotic-spectral-bound | AI | Apache-2.0 | 895.0 KB | not yet graded | an agent formalizing forbidden-subgraph spectral bounds in Lean queries this for general clique removal, weighted deletion, hub separation and uniform two-eigen |
| kkp-connected-spectral-counterexamples | shaikidris/kkp-connected-spectral-counterexamples | AI | Apache-2.0 | 156.6 KB | not yet graded | an agent formalizing spectral counterexamples in Lean queries this for the connected tripartite-bridge-K4 family, K5-minus-edge exclusion and exact two-vector a |
| kkp-localized-spectral-bound | shaikidris/kkp-localized-spectral-bound | AI | Apache-2.0 | 648.6 KB | not yet graded | an agent formalizing spectral graph inequalities in Lean queries this for the edge-clique scalar certificate, completely positive spectral transfer and localize |
| knox | pinterest/knox | CRYPTO | Apache-2.0 | 453.6 KB | not yet graded | An agent building key management and rotation queries this for Pinterest Knox's key store, ACLs and rotation model. |
| kotaemon | Cinnamon/kotaemon | AI | Apache-2.0 | 17.9 MB | not yet graded | An agent shipping a document-QA UI queries this for a clean RAG app architecture with reasoning pipelines and citations. |
| kratix | syntasso/kratix | OPS | Apache-2.0 | 5.2 MB | not yet graded | An agent building platform-as-a-product queries this for the Promise abstraction and request/fulfilment pipeline model. |
| kryptology | coinbase/kryptology | CRYPTO | Apache-2.0 | 7.0 MB | not yet graded | an agent working on threshold signatures/MPC or advanced primitives queries this for Coinbase's VSS, threshold ECDSA, BLS and accumulators |
| kube-bench | aquasecurity/kube-bench | SEC | Apache-2.0 | 4.6 MB | not yet graded | security |
| kube-hunter | aquasecurity/kube-hunter | SEC | Apache-2.0 | 729.9 KB | not yet graded | an agent probing a cluster queries this for kube-hunter's Kubernetes attack knowledge base (KB IDs) and test definitions. |
| kubernetes-community | kubernetes/community | API | Apache-2.0 | 58.8 MB | not yet graded | An agent designing declarative resource APIs (spec/status, list/watch, server-side apply, versioned groups) queries this for the Kubernetes API conventions and |
| kubescape-regolibrary | kubescape/regolibrary | SEC | Apache-2.0 | 6.2 MB | not yet graded | an agent assessing Kubernetes compliance queries this for Kubescape's Rego controls and their framework mappings (NSA, MITRE, CIS). |
| kyverno-policies | kyverno/policies | SEC | Apache-2.0 | 5.2 MB | not yet graded | an agent securing Kubernetes queries this for Kyverno's YAML policy library (pod security, best practices) and test resources. |
| lambdaworks | lambdaclass/lambdaworks | CRYPTO | Apache-2.0 | 6.7 MB | not yet graded | an agent working on STARK/SNARK proving math queries this for a from-scratch, well-documented Rust library of fields, FFTs and provers |
| ldnoobw-bad-words | LDNOOBW/List-of-Dirty-Naughty-Obscene-and-Otherwise-Bad-Words | WRITE | CC-BY-4.0 | 48.0 KB | not yet graded | An agent building profanity filters or content-safety checks queries this for obscene/offensive word lists across dozens of languages. |
| lego | go-acme/lego | CRYPTO | MIT | 6.4 MB | not yet graded | An agent automating ACME certificate issuance queries this for lego's ACME v2 client and DNS-01/HTTP-01/TLS-ALPN challenge providers. |
| leveldb | google/leveldb | DB | BSD-3-Clause | 969.1 KB | G7 4.0/5 | an agent learning LSM internals queries this for the original clean implementation (memtable/sstable/manifest split) |
| lexbuild | chris-c-thomas/lexbuild | LAW | MIT | 3.4 MB | not yet graded | An agent converting official legislative XML (USLM) into LLM-friendly structured Markdown queries this for a reference CLI and its transformation/metadata pipel |
| libcbor | PJK/libcbor | NET | MIT | 1.1 MB | not yet graded | NET |
| libhydrogen | jedisct1/libhydrogen | CRYPTO | ISC | 183.4 KB | not yet graded | An agent working on crypto for constrained/embedded targets queries this for libhydrogen's small, hard-to-misuse C API covering hashing, AEAD, signatures and ke |
| libopenapi | pb33f/libopenapi | API | MIT | 26.4 MB | not yet graded | An agent parsing and mutating OpenAPI 2/3 documents with full $ref resolution and a rolling index queries this for libopenapi's low/high model. |
| libsodium | jedisct1/libsodium | CRYPTO | ISC | 7.0 MB | not yet graded | An agent working on a modern crypto library (AEAD, X25519, Ed25519, secretstream, password hashing) queries this for libsodium's battle-tested constant-time C i |
| libuv | libuv/libuv | NET | MIT | 4.5 MB | not yet graded | an agent building an event loop queries this for the polling primitives and async threading patterns |
| licensee | licensee/licensee | LAW | MIT | 2.2 MB | not yet graded | An agent auto-detecting a repo's license from its LICENSE file queries this for the reference detection algorithm and the normalized license corpus GitHub itsel |
| lightningcss | parcel-bundler/lightningcss | WEB | MPL-2.0 | 10.7 MB | not yet graded | A Rust agent parsing/transforming CSS queries this for Lightning CSS's Rust CSS parser, minifier and vendor-prefixing/transpilation design. |
| lit | lit/lit | WEB | BSD-3-Clause | 12.8 MB | not yet graded | An agent building web components queries this for Lit's reactive-element lifecycle, declarative template directives and efficient DOM updates. |
| llhttp | nodejs/llhttp | NET | MIT | 7.8 MB | not yet graded | NET |
| llm-course | mlabonne/llm-course | AI | Apache-2.0 | 847.5 KB | not yet graded | an agent building with LLMs queries this for the fine-tune, quantize, RAG and deployment path and the concept behind each step |
| lmdb | LMDB/lmdb | DB | OpenLDAP-2.8 | 672.3 KB | G7 3.5/5 | an agent working on memory-mapped B+trees queries this for LMDB's page/cursor/COW mechanics |
| lokdocsqr | badretd/lokdocsqr | OPS | Apache-2.0 | 3.1 MB | not yet graded | an agent building or maintaining a local-first file-automation engine queries this for workflow-graph execution, scope-enforced filesystem mutations, dry-run an |
| lol-html | cloudflare/lol-html | WEB | BSD-3-Clause | 5.7 MB | not yet graded | A Rust agent rewriting HTML in a stream queries this for Cloudflare's low-latency streaming HTML rewriter and CSS-selector-driven rewriting. |
| loldrivers | magicsword-io/LOLDrivers | SEC | Apache-2.0 | 114.6 MB | not yet graded | an agent detecting BYOVD attacks queries this for LOLDrivers' known vulnerable/malicious driver metadata, hashes and YARA rules. |
| lz4 | lz4/lz4 | NET | mixed: BSD-2 (lib) / GPL-2.0+ (rest) | 1.6 MB | not yet graded | an agent needing fast block compression queries this for the block format and decoder loop |
| markdownlint | DavidAnson/markdownlint | WRITE | MIT | 3.7 MB | not yet graded | An agent authoring Markdown docs queries this for the consensus rule set on heading, list, spacing and formatting style (rules + rationale). |
| mashqa | mingzhu0527/MASHQA | HEALTH | Apache-2.0 | 664.9 KB | not yet graded | An agent building long-document consumer-health QA with multiple non-contiguous answer spans queries this for MASH-QA. |
| material-components-web | material-components/material-components-web | WEB | MIT | 10.7 MB | not yet graded | An agent implementing Material Design queries this for MDC component behaviour, ripple/elevation mechanics and the Sass theming system. |
| material-web | material-components/material-web | WEB | Apache-2.0 | 17.6 MB | not yet graded | An agent building Material 3 web components queries this for Google's MWC element implementations, theming, and ripple/motion behaviour. |
| mbedtls | Mbed-TLS/mbedtls | CRYPTO | Apache-2.0 OR GPL-2.0-or-later | 7.9 MB | not yet graded | An agent working on embedding or implementing a TLS stack queries this for mbed-TLS's portable C TLS 1.2/1.3 handshake, X.509 and crypto modules. |
| mdn-browser-compat-data | mdn/browser-compat-data | WEB | CC0-1.0 | 28.4 MB | not yet graded | an agent checking API/CSS/HTML/JS availability queries this for machine-readable per-feature browser-version compatibility records. |
| mdn-data | mdn/data | WEB | CC0-1.0 | 879.3 KB | not yet graded | An agent needing structured CSS/JS metadata (properties, at-rules, selectors, syntaxes, API inventory) queries this for MDN's machine-readable data. |
| medical-abbreviations | imantsm/medical_abbreviations | HEALTH | MIT | 175.6 KB | not yet graded | health |
| medmcqa | medmcqa/medmcqa | HEALTH | MIT | 750.7 KB | not yet graded | An agent building medical multiple-choice/exam QA queries this for ~194k Indian medical-entrance (AIIMS/NEET-PG) questions with explanations across 21 subjects. |
| medqa-usmle | jind11/MedQA | HEALTH | MIT | 100.7 KB | not yet graded | An agent building USMLE-style medical board QA queries this for MedQA multiple-choice questions (US/Mainland/Taiwan) plus the textbook passages used as context. |
| medquad | abachaa/MedQuAD | HEALTH | CC-BY-4.0 | 35.0 MB | not yet graded | health |
| medspacy | medspacy/medspacy | HEALTH | MIT | 5.3 MB | not yet graded | An agent building clinical-note NLP (sections, ConText negation/assertion, rule-based targets) queries this for medspaCy's clinical pipeline components. |
| mesh-patterns | autobanco-frd/mesh-patterns | AI | MIT | 30.8 KB | not yet graded | an agent building paid, verifiable services queries this for production patterns: x402 payment verification, A2A inboxes, Ed25519 signed receipts, deterministic |
| microsoft-seal | microsoft/SEAL | CRYPTO | MIT | 4.4 MB | not yet graded | an agent working on homomorphic encryption queries this for Microsoft SEAL's BFV/BGV/CKKS schemes and parameter selection |
| microsoft-style-vale | vale-cli/Microsoft | WRITE | MIT | 113.2 KB | not yet graded | an agent writing UI or docs copy queries this for Microsoft Writing Style Guide rules: terminology, tone, inclusive language, capitalization |
| miekg-dns | miekg/dns | NET | BSD-3-Clause | 912.3 KB | not yet graded | NET |
| milvus | milvus-io/milvus | DB | Apache-2.0 | 126.9 MB | not yet graded | DB |
| minify | tdewolff/minify | WEB | MIT | 37.9 MB | not yet graded | An agent minifying web assets server-side queries this for tdewolff's HTML/CSS/JS/SVG/JSON minifiers and their tokenizers. |
| minisign | jedisct1/minisign | CRYPTO | ISC | 90.7 KB | not yet graded | an agent working on simple Ed25519 file signing/verification queries this for minisign's compact signature format and trusted-comment scheme |
| misp-galaxy | MISP/misp-galaxy | SEC | CC0-1.0 OR BSD-2-Clause | 57.6 MB | not yet graded | security |
| mitre-attck | xerj-org/corpus-mitre-attck | SEC | MITRE ATT&CK Terms of Use (permissive, notice-preserving) | 1.8 MB | G7 5.0/5 | an agent writing or reviewing detections queries this for technique ids, tactics, detection data-sources and procedure examples |
| mitre-capec | xerj-org/corpus-mitre-capec | SEC | MITRE Terms of Use (permissive, notice-preserving) | 824.4 KB | G7 3.0/5 | an agent reasoning about attack surface queries this for attack-pattern enumerations with prerequisites and mitigations |
| mitre-cwe | xerj-org/corpus-mitre-cwe | SEC | MITRE Terms of Use (permissive, notice-preserving) | 2.1 MB | G7 3.0/5 | an agent classifying a finding queries this for the CWE definition, its parents/children and detection notes |
| mkcert | FiloSottile/mkcert | CRYPTO | BSD-3-Clause | 54.5 KB | not yet graded | An agent working on locally-trusted development TLS certificates queries this for mkcert's local-CA creation and system/browser trust-store installation logic. |
| mls-rs | awslabs/mls-rs | CRYPTO | Apache-2.0 | 61.6 MB | not yet graded | An agent implementing MLS in Rust queries this for AWS Labs' mls-rs, an alternative RFC 9420 implementation with pluggable crypto providers. |
| monocypher | LoupVaillant/Monocypher | CRYPTO | BSD-2-Clause OR CC0-1.0 | 1.3 MB | not yet graded | An agent working on a tiny auditable crypto library queries this for Monocypher's single-file C implementations of X25519, EdDSA, ChaCha20, BLAKE2 and Argon2. |
| mozilla-diversity | mozilla/diversity | LAW | MPL-2.0 | 308.5 KB | not yet graded | An agent drafting community participation guidelines or a code of conduct with an enforcement process queries this for Mozilla's CPG and CoC templates. |
| mozilla-l10n-docs | mozilla-l10n/localizer-documentation | WRITE | CC-BY-4.0 | 1023.5 KB | not yet graded | An agent writing localization/translation style guidance queries this for Mozilla's localizer documentation and per-locale translation style conventions. |
| mozilla-l10n-styleguides | mozilla-l10n/styleguides | WRITE | CC-BY-4.0 | 2.7 MB | not yet graded | An agent localising software queries this for Mozilla's per-locale tone, punctuation and terminology style guides. |
| mozilla-ssl-configs | mozilla/server-side-tls | CRYPTO | MPL-2.0 | 37.1 KB | G7 3.0/5 | an agent configuring TLS termination queries this for the recommended cipher/protocol sets per server and their security/cost tradeoffs |
| moztw-styleguides | moztw/styleguides | WRITE | CC-BY-4.0 | 2.6 MB | not yet graded | An agent translating software UI/content queries this for community translation style guides (tone, terminology, formatting) across locales. |
| msft-api-guidelines | microsoft/api-guidelines | API | CC-BY-4.0 | 611.5 KB | G7 4.0/5 | an agent designing a REST surface queries this for the naming/paging/versioning/error rules with examples |
| msgpack-c | msgpack/msgpack-c | NET | BSL-1.0 | 28.7 KB | not yet graded | an agent implementing msgpack queries this for the format handling and unpack edge cases |
| mtl-bioinformatics-ner | cambridgeltl/MTL-Bioinformatics-2016 | HEALTH | CC-BY-4.0 | 212.2 MB | not yet graded | An agent training biomedical named-entity recognition queries this for a bundle of CoNLL-format corpora (NCBI-Disease, BC5CDR, BioNLP, JNLPBA, Linnaeus, Species |
| mts-dialog-clinical-notes | abachaa/MTS-Dialog | HEALTH | CC-BY-4.0 | 7.9 MB | not yet graded | health |
| mutcd | xerj-org/corpus-mutcd | LAW | US-government public domain (FHWA/DOT, 17 USC 105) | 33.3 MB | G7 4.0/5 | the current MUTCD: sign and signal warrants, markings, work-zone typical applications |
| nats-protocol | nats-io/nats-architecture-and-design | DB | Apache-2.0 | 968.5 KB | G7 4.0/5 | an agent implementing a NATS client queries this for the wire protocol, verbs and semantics text |
| nats-server | nats-io/nats-server | DB | Apache-2.0 | 14.1 MB | not yet graded | an agent working on pub/sub queries this for subject-tree routing, jetstream and clustering internals in readable Go |
| navrules | xerj-org/corpus-navrules | LAW | US-government public domain (USCG) | 270.5 KB | G7 4.5/5 | a mariner, boater, or agent writing navigation software queries this for the actual text of Rule 7 risk of collision, Rule 9 narrow channels, lights, sound sign |
| nghttp2 | nghttp2/nghttp2 | NET | MIT | 5.5 MB | not yet graded | an agent working with HTTP/2 queries this for flow control, HPACK and priority internals in the reference C library |
| nginx | nginx/nginx | NET | BSD-2-Clause | 7.9 MB | not yet graded | an agent working on web servers queries this for the event-driven request cycle and module seams (note: very large) |
| nigeria-standard-treatment-guidelines | Aybee5/nigeria-clinical-guidelines-dataset | HEALTH | MIT | 1.7 MB | not yet graded | health |
| nuclei-templates | projectdiscovery/nuclei-templates | SEC | MIT | 43.9 MB | not yet graded | security |
| oasdiff | oasdiff/oasdiff | API | Apache-2.0 | 5.9 MB | not yet graded | An agent detecting breaking changes between OpenAPI versions queries this for oasdiff's diff algorithm and breaking-change classification. |
| openapi-moonwalk | OAI/sig-moonwalk | API | Apache-2.0 | 178.2 KB | not yet graded | An agent anticipating the next-generation OpenAPI data model queries this for the Moonwalk SIG proposals and design discussions. |
| openapi-overlay-spec | OAI/Overlay-Specification | API | Apache-2.0 | 316.5 KB | not yet graded | An agent programmatically patching/transforming OpenAPI documents queries this for the Overlay actions (target/update/remove) format. |
| openapi-spec | OAI/OpenAPI-Specification | API | Apache-2.0 | 2.4 MB | G7 5.0/5 | an agent generating or validating an API description queries this for the OpenAPI schema's exact keywords and resolution rules |
| openbao | openbao/openbao | CRYPTO | MPL-2.0 | 72.5 MB | not yet graded | an agent working on secret stores queries this for the barrier, seal and lease internals of the open vault fork |
| openbsd-faq | openbsd/www | OPS | ISC-style per-chapter grants | 760.8 MB | G7 3.0/5 | an agent hardening an OpenBSD host queries this for the pf FAQ's rule grammar, upgrade procedures and faq-page hardening steps |
| openfda | FDA/openfda | HEALTH | CC0-1.0 | 156.7 MB | not yet graded | An agent working on drug- and device-safety questions queries this for FDA adverse-event (FAERS), recall/enforcement, NDC, and device data field schemas and har |
| openlegaldata-oldp | openlegaldata/oldp | LAW | MIT | 28.0 MB | not yet graded | An agent building a platform that ingests and serves court decisions and laws queries this for the Open Legal Data Platform (Django) data model and legal-refere |
| opennyai-legal-nlp | OpenNyAI/Opennyai | LAW | MIT | 3.7 MB | not yet graded | An agent building Indian legal NLP (judgment structuring, legal NER, rhetorical-role tagging) queries this for a spaCy pipeline trained on Indian court text. |
| opensanctions | opensanctions/opensanctions | LAW | MIT | 28.7 MB | not yet graded | An agent building sanctions/PEP screening and KYC compliance queries this for consolidated sanctions-list data models and ETL across OFAC/EU/UN sources. |
| openssl | openssl/openssl | CRYPTO | Apache-2.0 | 132.4 MB | not yet graded | an agent debugging TLS/ASN.1 queries this for the reference provider's handshake, BIO and provider internals |
| osquery-config-examples | palantir/osquery-configuration | SEC | MIT | 214.6 KB | G7 4.0/5 | an agent writing host detections queries this for battle-tested osquery SQL by objective (suid diffs, persistence paths, launch items) |
| oss-attribution-builder | amzn/oss-attribution-builder | LAW | Apache-2.0 | 1.3 MB | not yet graded | An agent generating third-party attribution/notice documents queries this for the workflow and license-text templates that assemble an OSS attribution document. |
| oss-license-texts | github/choosealicense.com | LAW | MIT | 1.1 MB | not yet graded | law |
| oss-review-toolkit | oss-review-toolkit/ort | LAW | Apache-2.0 | 78.6 MB | not yet graded | An agent building a dependency license-compliance pipeline (SBOM, policy, attribution) queries this for a production OSS Review Toolkit implementation. |
| otel-proto | open-telemetry/opentelemetry-proto, xerj-org/corpus-otel-semconv-docs | OPS | Apache-2.0+Apache-2.0 (mirror LICENSE) / CC-BY-4.0 (mirrored docs) | 4.7 MB | G7 0.0/5 | an agent wiring telemetry queries this for the exact semantic conventions and proto field semantics |
| owasp-cheatsheets | OWASP/CheatSheetSeries | SEC | CC-BY-SA-4.0 | 23.0 MB | G7 5.0/5 | an agent implementing an auth/upload/crypto feature queries this for the concrete cheatsheet: header lists, code shapes, test ideas |
| owasp-wstg | OWASP/wstg | SEC | CC-BY-SA-4.0 | 14.6 MB | G7 5.0/5 | an agent planning a web test pass queries this for test categories, procedures and expected results |
| oxc | oxc-project/oxc | WEB | MIT | 75.1 MB | not yet graded | A Rust agent building fast JS tooling queries this for Oxc's Rust parser, linter and resolver - a modern, performance-focused reference. |
| oxidize-pdf | bzsanti/oxidizePdf | WRITE | MIT AND Apache-2.0 AND BSD-3-Clause AND LicenseRef-Adobe-AFM | 16.3 MB | not yet graded | An agent implementing PDF workflows in Rust queries this for oxidize-pdf document generation, PDF parsing, text extraction, font encodings and encrypted object |
| pagerduty-incident-response | PagerDuty/incident-response-docs | OPS | Apache-2.0 | 170.4 MB | not yet graded | An agent writing on-call runbooks or incident-response automation queries this for severity levels, responder roles, and postmortem process. |
| parquet-format | apache/parquet-format | DB | Apache-2.0 | 895.9 KB | not yet graded | an agent reading/writing parquet queries this for encodings, logical types and the thrift metadata spec |
| parse5 | inikulin/parse5 | WEB | MIT | 3.5 MB | not yet graded | An agent parsing real-world HTML queries this for a spec-compliant HTML5 tree construction/tokenizer in JS, including error recovery. |
| payloadsallthethings | swisskyrepo/PayloadsAllTheThings | SEC | MIT | 12.4 MB | not yet graded | security |
| pebble | cockroachdb/pebble | DB | BSD | 20.0 MB | G7 4.0/5 | an agent working on LSM stores queries this for a modern Go LSM with delete/RangeLock semantics |
| plain-language-guidelines | GSA/plainlanguage.gov | WRITE | CC0-1.0 | 64.8 MB | not yet graded | An agent rewriting bureaucratic or legalese text into plain language queries this for the U.S. Federal Plain Language Guidelines and their before/after examples |
| plan-and-act | kamwoh/plan-and-act | AI | MIT | 3.6 MB | not yet graded | an agent building an LLM that plans and controls a robot arm queries this for how plan-and-act has Claude imagine a plan as timed 3D keyframes, run it on a MuJo |
| polars | pola-rs/polars | DB | MIT | 30.0 MB | not yet graded | DB |
| postcss | postcss/postcss | WEB | MIT | 730.0 KB | not yet graded | An agent transforming CSS queries this for PostCSS's CSS parser, node/AST model and visitor/plugin pipeline. |
| postgres-src | postgres/postgres | DB | PostgreSQL | 139.9 MB | not yet graded | an agent working on planners queries this for the real cost model, join ordering and executor nodes |
| postman-collection | postmanlabs/postman-collection | API | Apache-2.0 | 1.6 MB | not yet graded | An agent reading/writing Postman Collection v2 documents queries this SDK for the collection/request/variable object model. |
| preact | preactjs/preact | WEB | MIT | 1.6 MB | not yet graded | An agent wanting a compact React-compatible VDOM queries this for Preact's small diff/reconciler and hooks implementation. |
| prism | stoplightio/prism | API | Apache-2.0 | 2.7 MB | not yet graded | An agent building an HTTP mock/validation proxy from an OpenAPI document queries this for Prism's request validation and example/dynamic response generation. |
| prometheus-src | prometheus/prometheus | DB | Apache-2.0 | 28.3 MB | not yet graded | an agent working on TSDB queries this for the head-block/WAL/head-compaction paths and PromQL evaluation |
| prompt-engineering-guide | dair-ai/Prompt-Engineering-Guide | AI | MIT | 76.2 MB | not yet graded | an agent applying a prompting technique queries this for CoT, ReAct, few-shot and RAG-prompting definitions and worked examples |
| promptify | promptslab/Promptify | AI | Apache-2.0 | 2.1 MB | not yet graded | An agent doing prompt-based NLP (NER, classification, QA) without fine-tuning queries this for Promptify's task-prompt templates and output parsers. |
| proselint | amperser/proselint | WRITE | BSD-3-Clause | 2.7 MB | not yet graded | an agent reviewing prose queries this for sourced style checks: redundancy, jargon, hedging, misused phrases, each with a citation |
| protobuf | protocolbuffers/protobuf | NET | BSD-3-Clause | 99.0 MB | not yet graded | an agent generating or debugging protobuf code queries this for wire format, descriptor handling and the conformance suite |
| protovalidate | bufbuild/protovalidate | API | Apache-2.0 | 3.6 MB | not yet graded | An agent expressing field/message validation constraints on protobuf APIs queries this for protovalidate's CEL-based constraint annotations. |
| pubhealth-fact-checking | neemakot/Health-Fact-Checking | HEALTH | MIT | 185.3 KB | not yet graded | An agent building public-health claim verification queries this for the PUBHEALTH dataset: 11.8k claims with veracity labels and journalist-written gold explana |
| pubmedqa | pubmedqa/pubmedqa | HEALTH | MIT | 2.5 MB | not yet graded | An agent building biomedical research yes/no/maybe question answering queries this for expert-labeled PubMed-abstract QA pairs plus unlabeled and generated spli |
| pysec | xerj-org/corpus-pysec-v2 | SEC | CC-BY-4.0 | 13.5 MB | G7 4.5/5 | an agent reviewing Python dependency trees queries this for PYSEC advisory records with fixed versions |
| pyverge | JoHa-HQ/pyverge | API | Apache-2.0 | 2.5 MB | not yet graded | an agent designing schema-evolution runtimes queries this for how pyverge models version graphs, migration paths, target policies, reflection, and registry-back |
| quiche | cloudflare/quiche | NET | BSD-2-Clause | 9.5 MB | not yet graded | NET |
| quinn | quinn-rs/quinn | NET | Apache-2.0 OR MIT | 2.4 MB | G7 4.0/5 | an agent implementing QUIC queries this for congestion control, stream state and 0-RTT handling in idiomatic Rust |
| qwik | QwikDev/qwik | WEB | MIT | 36.2 MB | not yet graded | An agent minimising client JS queries this for Qwik's resumability model and code-splitting/lazy-loading of event handlers. |
| r2r-enforce | Thneoly/r2r-jev | AI | Apache-2.0 | 280.5 KB | not yet graded | an agent closing the decision/execution gap queries this for how r2r-enforce binds an execution to the governance decision and state version that authorized it |
| r2r-mcp | Thneoly/r2r-jev | AI | Apache-2.0 | 214.7 KB | not yet graded | an agent wiring governance into an MCP client queries this for how r2r-mcp exposes observe, decide and explain over R2R relation state on stdio |
| rabbitmq | rabbitmq/rabbitmq-server | DB | MPL-2.0 (core) AND Apache-2.0 (OCF parts) | 60.4 MB | not yet graded | an agent working on message brokers queries this for queue/exchange routing and mirrored-queue semantics |
| raft-rs | tikv/raft-rs | DB | Apache-2.0 | 922.4 KB | not yet graded | an agent implementing raft queries this for the log/ready/progress state machine the tikv fleet runs on |
| rctool-live2d | richuff/RcTool | WEB | MIT AND LicenseRef-Live2D-Proprietary | 8.7 MB | not yet graded | An agent embedding a Cubism Live2D character in a Flutter app queries this for WebView model loading through a local HTTP asset server, Cubism 4 lip sync (Param |
| react-spectrum | adobe/react-spectrum | WEB | Apache-2.0 | 187.7 MB | not yet graded | An agent building accessible React components queries this for React Aria/React Stately implementations of ARIA design patterns, focus management, and i18n. |
| redhat-vale-style | redhat-documentation/vale-at-red-hat | WRITE | MIT | 39.7 MB | not yet graded | An agent applying an enterprise technical-writing style queries this for Red Hat's supplementary Vale rules and word-usage guidance. |
| redoc | Redocly/redoc | API | MIT | 33.0 MB | not yet graded | An agent rendering OpenAPI as reference documentation queries this for how Redoc resolves/normalises specs and lays out three-panel docs. |
| redocly-cli | Redocly/redocly-cli | API | MIT | 23.9 MB | not yet graded | An agent bundling, splitting and linting OpenAPI definitions queries this for Redocly CLI's config, built-in rules and bundling logic. |
| reporters-db | freelawproject/reporters-db | LAW | BSD-2-Clause | 1.7 MB | not yet graded | An agent parsing or validating U.S. legal citations queries this for the structured database of court reporter abbreviations, editions, and variations. |
| retext-equality | retextjs/retext-equality | WRITE | MIT | 172.3 KB | not yet graded | An agent correcting gendered, ableist, or polarising language queries this for the exact term-to-suggestion rule data (equality.yml) that powers alex. |
| retext-indefinite-article | retextjs/retext-indefinite-article | WRITE | MIT | 22.2 KB | not yet graded | An agent copy-editing English queries this for a/an indefinite-article correctness rules including exceptions like 'an hour' and 'a user'. |
| retext-intensify | retextjs/retext-intensify | WRITE | MIT | 15.8 KB | not yet graded | An agent tightening prose queries this for weak filler intensifiers and hedges to flag (e.g. 'very', 'really', 'quite'). |
| retext-passive | retextjs/retext-passive | WRITE | MIT | 16.6 KB | not yet graded | An agent enforcing active voice queries this for the word list and approach used to detect passive-voice constructions. |
| retext-profanities | retextjs/retext-profanities | WRITE | MIT | 22.3 KB | not yet graded | An agent moderating or sanitizing copy queries this for profanity term lists with severity ratings and suggested rephrasings. |
| retext-quotes | retextjs/retext-quotes | WRITE | MIT | 30.7 KB | not yet graded | An agent normalizing typography in prose queries this for straight-vs-curly quote and apostrophe rules. |
| retext-readability | retextjs/retext-readability | WRITE | MIT | 24.2 KB | not yet graded | An agent assessing text difficulty queries this for how Dale-Chall/ARI/Coleman-Liau/Flesch/Gunning-Fog/SMOG/Spache are combined to flag hard sentences. |
| retext-redundant-acronyms | retextjs/retext-redundant-acronyms | WRITE | MIT | 21.9 KB | not yet graded | An agent editing technical copy queries this for the list of redundant acronym expansions ('ATM machine','PIN number'). |
| retext-repeated-words | retextjs/retext-repeated-words | WRITE | MIT | 16.9 KB | not yet graded | An agent proofreading queries this for the rule that catches accidental repeated words ('the the'). |
| retext-simplify | retextjs/retext-simplify | WRITE | MIT | 32.4 KB | not yet graded | An agent writing plain-language prose queries this for the list of complex/wordy phrases and their simpler replacements. |
| rocksdb | facebook/rocksdb | DB | GPL-2.0 OR Apache-2.0 (Apache selected) | 48.5 MB | not yet graded | an agent working on LSM stores queries this for compaction, WAL and recovery internals at production scale |
| rolldown | rolldown/rolldown | WEB | MIT | 26.1 MB | not yet graded | A Rust agent building a bundler queries this for Rolldown's Rust module graph and Rollup-compatible bundling design. |
| rousseau | GitbookIO/rousseau | WRITE | Apache-2.0 | 151.6 KB | not yet graded | An agent proofreading English prose queries this for lightweight checks: weasel words, passive voice, cliches, readability and lexical illusions. |
| rspack | web-infra-dev/rspack | WEB | MIT | 66.6 MB | not yet graded | An agent building a high-performance bundler queries this for a Rust reimplementation of webpack's module/loader/plugin/HMR pipeline. |
| rust-api-guidelines | rust-lang/api-guidelines | API | MIT OR Apache-2.0 | 106.7 KB | G7 5.0/5 | an agent designing a Rust library API queries this for the naming/convertibility/error conventions with their reasoning |
| rust-cssparser | servo/rust-cssparser | WEB | MPL-2.0 | 1.9 MB | not yet graded | A Rust agent tokenising CSS queries this for Servo's low-level CSS tokenizer and parser-combinator design. |
| rust-url | servo/rust-url | WEB | Apache-2.0 | 2.4 MB | not yet graded | A Rust agent parsing/normalising URLs queries this for the `url` crate - a Rust implementation of the WHATWG URL Standard (parsing, IDNA, origin). |
| rust-vulns | rustsec/advisory-db | SEC | CC0-1.0 | 0 B | not yet graded | an agent reviewing Rust code queries this for vulnerability precedent: identity-resolved RUSTSEC/GHSA/CVE advisories with affected functions and fixed versions |
| rustcrypto-hashes | RustCrypto/hashes | CRYPTO | Apache-2.0 OR MIT | 3.9 MB | not yet graded | An agent working on hash-function implementations queries this for RustCrypto's pure-Rust SHA-2/SHA-3/BLAKE2/Keccak code and the shared digest-trait design. |
| rustls | rustls/rustls | CRYPTO | Apache-2.0 OR MIT | 12.5 MB | not yet graded | an agent implementing TLS in Rust queries this for the state machine, cipher dispatch and certificate verification |
| s2n-tls | aws/s2n-tls | CRYPTO | Apache-2.0 | 21.4 MB | not yet graded | an agent hardening TLS queries this for the defensively-structured C state machine and its testing approach |
| sass-spec | sass/sass | WEB | MIT | 1.7 MB | not yet graded | An agent implementing a CSS preprocessor queries this for the Sass language specification, module system, and evaluation semantics. |
| scap-security-guide | ComplianceAsCode/content | SEC | BSD-3-Clause | 65.2 MB | not yet graded | security |
| school-of-sre | linkedin/school-of-sre | OPS | CC-BY-4.0 | 20.8 MB | G7 5.0/5 | an agent stepping into an SRE task queries this for the curriculum's concrete replication/signals/messaging/git/networking fundamentals (no incident-response or |
| scispacy | allenai/scispacy | HEALTH | Apache-2.0 | 588.3 KB | not yet graded | An agent doing scientific/biomedical text processing in spaCy queries this for scispaCy's tokenization, abbreviation detection, NER and entity-linking code. |
| season-of-docs | google/season-of-docs | WRITE | CC-BY-4.0 | 821.4 KB | not yet graded | An agent planning a technical-writing project queries this for Google Season of Docs guidance on scoping, producing and measuring docs. |
| serverless-workflow-spec | serverlessworkflow/specification | API | Apache-2.0 | 14.7 MB | not yet graded | An agent designing declarative workflow/orchestration APIs (states, events, retries, function refs) queries this for the CNCF Serverless Workflow DSL. |
| simonw-llm | simonw/llm | AI | Apache-2.0 | 1.9 MB | not yet graded | An agent building an LLM CLI, logging model calls, or writing a model plugin queries this for the `llm` tool's architecture and its extensive, well-written docs |
| siren-hypermedia | kevinswiber/siren | API | MIT | 26.7 KB | not yet graded | An agent designing a hypermedia JSON format with entities, actions and links queries this for the Siren specification. |
| smallstep-certificates | smallstep/certificates | CRYPTO | Apache-2.0 | 4.7 MB | not yet graded | An agent working on a private CA, ACME server or certificate-automation workflow queries this for step-ca's X.509/SSH issuance, provisioners and ACME implementa |
| smithy | smithy-lang/smithy | API | Apache-2.0 | 21.0 MB | not yet graded | An agent modelling protocol-agnostic service APIs (shapes, traits, protocol bindings) queries this for the AWS Smithy IDL and model semantics. |
| snappy | google/snappy | NET | BSD-3-Clause | 3.2 MB | not yet graded | an agent needing cheap compression queries this for the format and the reference encoder decisions |
| solid | solidjs/solid | WEB | MIT | 1.3 MB | not yet graded | An agent building fine-grained reactivity queries this for SolidJS's signal/effect graph and compile-time JSX reactivity. |
| sourcegraph-handbook | sourcegraph/handbook | WRITE | Apache-2.0 | 9.1 MB | not yet graded | An agent writing an open company handbook queries this for a large real-world example of handbook information architecture and plainspoken internal-doc writing. |
| spdx-spec | spdx/spdx-spec | LAW | CC-BY-3.0 | 17.6 MB | not yet graded | An agent implementing SBOM generation or license declaration queries this for the SPDX specification prose (document structure, relationships, license-expressio |
| spectral | stoplightio/spectral | API | Apache-2.0 | 7.1 MB | not yet graded | An agent linting OpenAPI/AsyncAPI documents queries this for Spectral's rule/function engine and the built-in OAS style rulesets. |
| spectrum-web-components | adobe/spectrum-web-components | WEB | Apache-2.0 | 42.4 MB | not yet graded | An agent building against Adobe's Spectrum queries this for Spectrum web-component APIs, theming and token usage. |
| splunk-security-content | splunk/security_content | SEC | Apache-2.0 | 38.3 MB | not yet graded | security |
| sqlite-src | sqlite/sqlite | DB | Public-Domain (SQLite blessing) | 107.8 MB | not yet graded | an agent working on embedded DB internals queries this for the btree module, pager and the famously commented test harness |
| standard-readme | RichardLitt/standard-readme | WRITE | MIT | 69.5 KB | not yet graded | An agent writing README files queries this for the Standard Readme specification (required sections, ordering, wording) and generator. |
| standard-webhooks | standard-webhooks/standard-webhooks | API | Apache-2.0 | 363.2 KB | not yet graded | An agent implementing or consuming webhooks queries this for the Standard Webhooks spec covering payload signing, signature verification, idempotency, and deliv |
| stratus-red-team | DataDog/stratus-red-team | SEC | Apache-2.0 | 3.9 MB | not yet graded | security |
| stripe-openapi | stripe/openapi | API | MIT | 188.0 MB | not yet graded | An agent designing a large versioned payments/REST API queries this for how a production provider models resources, object expansion, cursor pagination and idem |
| style-dictionary | style-dictionary/style-dictionary | WEB | Apache-2.0 | 7.2 MB | not yet graded | An agent transforming design tokens into platform outputs (CSS/iOS/Android) queries this for Style Dictionary's parse/transform/format pipeline. |
| svelte | sveltejs/svelte | WEB | MIT | 6.9 MB | not yet graded | An agent building a reactive compiler/framework queries this for Svelte's compile-time reactivity model, runes and template compilation. |
| swagger-ui | swagger-api/swagger-ui | API | Apache-2.0 | 17.3 MB | not yet graded | An agent building an interactive API console queries this for Swagger UI's OpenAPI parsing and try-it-out request model. |
| swift-book | swiftlang/swift-book | API | Apache-2.0 | 8.7 MB | not yet graded | An agent writing Swift queries this for The Swift Programming Language book: syntax and semantics of concurrency (async/await, async let, actors), generics, opt |
| swift-evolution | swiftlang/swift-evolution | API | Apache-2.0 | 11.2 MB | not yet graded | An agent writing Swift queries this for the accepted Swift Evolution proposals (SE-NNNN): the motivation, exact semantics and API of language and standard-libra |
| symptom-ontology | DiseaseOntology/SymptomOntology | HEALTH | CC0-1.0 | 3.1 MB | not yet graded | An agent working on triaging or normalizing patient-reported complaints queries this for a controlled vocabulary of disease symptoms with definitions and synony |
| synthea | synthetichealth/synthea | HEALTH | Apache-2.0 | 121.9 MB | not yet graded | An agent building or testing EHR/FHIR pipelines queries this for Synthea's synthetic-patient generator: disease-progression modules, clinical logic and FHIR/C-C |
| texio | Allra-Fintech/texio | WRITE | MIT AND BSD-3-Clause | 574.7 KB | not yet graded | An agent inspecting or safely editing Markdown queries this for Texio heading selection, section extraction, dry-run replacements, ambiguity refusal, and tested |
| textlint | textlint/textlint | WRITE | MIT | 10.9 MB | not yet graded | An agent building a pluggable natural-language linter queries this for the engine design (rules, fixers, AST over Markdown/text) used across English and Japanes |
| textlint-ja-technical-writing | textlint-ja/textlint-rule-preset-ja-technical-writing | WRITE | MIT | 533.8 KB | not yet graded | An agent writing or linting Japanese technical documentation queries this preset for Japanese technical-writing rules (sentence length, doubled particles, max-t |
| textlint-rule-preset-ai-writing | textlint-ja/textlint-rule-preset-ai-writing | WRITE | MIT | 894.4 KB | not yet graded | An agent reviewing machine-generated text queries this for patterns characteristic of AI-written prose (formulaic hedging, list tics) to flag and correct. |
| textlint-rule-preset-jtf-style | textlint-ja/textlint-rule-preset-JTF-style | WRITE | MIT | 1.1 MB | not yet graded | An agent writing or localising Japanese content queries this for the JTF (Japan Translation Federation) standard style rules (full/half-width, punctuation, nume |
| textlint-rule-stop-words | sapegin/textlint-rule-stop-words | WRITE | MIT | 466.4 KB | not yet graded | An agent tightening prose queries this for a curated list of weasel/filler 'stop words' to remove. |
| textlint-rule-terminology | sapegin/textlint-rule-terminology | WRITE | MIT | 225.9 KB | not yet graded | An agent enforcing house terminology queries this for the canonical spelling and capitalisation of hundreds of tech and product terms (JavaScript, GitHub, npm.. |
| textstat | textstat/textstat | WRITE | MIT | 399.1 KB | not yet graded | An agent computing reading-level or readability scores queries this for implementations of Flesch, Flesch-Kincaid, Gunning-Fog, SMOG, Coleman-Liau, ARI and Dale |
| thrift | apache/thrift | NET | Apache-2.0 | 19.7 MB | not yet graded | NET |
| tikv | tikv/tikv | DB | Apache-2.0 | 40.3 MB | not yet graded | DB |
| tldr-pages | xerj-org/corpus-tldr-common | OPS | CC-BY-4.0 | 2.7 MB | G7 2.0/5 | an agent scripting an unfamiliar long-tail tool queries this for its example invocations (popular tools are memorised: G3 value is the tail only) |
| trivy-checks | aquasecurity/trivy-checks | SEC | MIT | 2.6 MB | not yet graded | security |
| twelve-factor | heroku/12factor | OPS | MIT | 21.8 MB | G7 5.0/5 | an agent containerising or configuring an app queries this for the factor's exact statement and rationale (G3 borderline: well-known prose: keep only if G7 show |
| uk-building-regs | xerj-org/corpus-uk-building-regs | LAW | NONE-FOUND | 522.7 KB | G7 5.0/5 | current UK Building Regulations 2010 text: what work must comply with, materials, loading, exemptions, notices |
| uk-legislation | xerj-org/corpus-uk-legislation | LAW | OGL-3.0 | 62.8 MB | G7 5.0/5 | an agent answering UK-law questions queries this for what the current statute actually says: Consumer Rights Act remedies, GDPR-adjacent DPA 2018 duties, Buildi |
| unicode-cldr | xerj-org/corpus-unicode-cldr | WEB | Unicode-3.0 | 4.3 MB | G7 5.0/5 | an agent formatting dates/numbers/names queries this for locale-specific CLDR rules and plural/data tables no model recalls |
| unitedstates-citation | unitedstates/citation | LAW | CC0-1.0 | 2.4 MB | not yet graded | An agent parsing references to US law (USC/CFR/public laws) queries this for a public-domain legal citation parser and grammar. |
| uslm-us-legislation | usgpo/uslm | LAW | CC0-1.0 | 166.1 MB | not yet graded | law |
| valkey-docs | valkey-io/valkey-doc | DB | CC-BY-SA-4.0 | 3.0 MB | G7 4.0/5 | an agent building against RESP queries this for command semantics, ranges and error cases (the open post-relicence fork of the Redis docs) |
| valkey-src | valkey-io/valkey | DB | BSD-3-Clause | 22.3 MB | not yet graded | an agent working on in-memory KV queries this for the event-loop, object encoding and replication paths of the open fork |
| vanguard-context-burn | VincentChuWaiChow/vanguard-frontier-agentic | AI | Apache-2.0 | 34.5 KB | not yet graded | an agent implementing coding-agent context tracking queries this for cross-harness context-budget hooks, usage accounting and regression tests |
| vectra-core | ronitgupta138/Vectra-Core | AI | MIT | 104.5 KB | not yet graded | an agent working on vector search and HNSW graph indexing queries this for how Vectra-Core implements Java 21 Loom virtual threads, cosine/euclidean distance me |
| vite | vitejs/vite | WEB | MIT | 16.9 MB | not yet graded | An agent building a dev server/bundler queries this for Vite's native-ESM dev server, dependency pre-bundling and plugin/HMR architecture. |
| vue | vuejs/core | WEB | MIT | 6.2 MB | not yet graded | An agent building a reactive runtime queries this for Vue 3's reactivity system (ref/reactive/effect), the virtual-DOM renderer and template compiler. |
| vuln-fix-commits | SAP/project-kb | SEC | Apache-2.0 | 383.7 MB | G7 4.0/5 | an agent confirming a vulnerability is actually fixed queries this for the fixing commit references (SAP project-kb cvefixcommit) |
| w3c-css | w3c/csswg-drafts | WEB | W3C-Software-And-Document | 141.6 MB | G7 4.0/5 | an agent working on layout/selector internals queries this for the csswg draft's algorithm text (the resolution and cascade algorithms are unmemorisable prose) |
| weaviate | weaviate/weaviate | DB | BSD-3-Clause AND LicenseRef-Weaviate-Enterprise | 54.2 MB | not yet graded | DB |
| web-features | web-platform-dx/web-features | WEB | Apache-2.0 | 4.0 MB | not yet graded | An agent checking Baseline availability of web platform features queries this for the web-features dataset that powers Baseline badges. |
| webgl | KhronosGroup/WebGL | WEB | MIT | 84.5 MB | not yet graded | An agent implementing or validating WebGL queries this for the WebGL1/2 IDL, spec text and the conformance test suite (expected behaviour per call). |
| whatwg-dom | whatwg/dom | WEB | CC-BY-4.0 | 7.3 MB | not yet graded | an agent manipulating page structure queries this for the authoritative DOM tree, event, mutation, and node-interface semantics (including Shadow DOM / custom e |
| whatwg-fetch | whatwg/fetch | WEB | CC-BY-4.0 | 6.1 MB | not yet graded | an agent writing browser networking code queries this for the Fetch Standard's request/response model, CORS, redirect, and caching semantics. |
| whatwg-html | whatwg/html | WEB | CC-BY-4.0 | 117.2 MB | not yet graded | an agent building or debugging web pages queries this for the HTML Living Standard's exact parsing algorithm, element semantics, and DOM-serialization rules. |
| whatwg-specs | xerj-org/corpus-whatwg-sections | WEB | CC-BY-4.0 (text) + BSD-3-Clause (code), per file | 7.8 MB | G7 0.0/5 | an agent building browser-adjacent tooling queries this for the living standard's exact normative statements (html/url/fetch/dom) |
| whatwg-streams | whatwg/streams | WEB | CC-BY-4.0 | 7.5 MB | not yet graded | an agent handling streaming I/O in the browser queries this for the Streams Standard's readable/writable/transform semantics and backpressure rules. |
| whatwg-url | whatwg/url | WEB | CC-BY-4.0 | 2.5 MB | not yet graded | an agent parsing or validating URLs queries this for the WHATWG URL Standard's exact parsing state machine, origin rules, and the URLSearchParams API. |
| whatwg-webidl | whatwg/webidl | WEB | CC-BY-4.0 | 7.2 MB | not yet graded | an agent implementing or calling web APIs queries this for Web IDL's type system and the binding rules every browser interface is defined with. |
| woke-inclusive-lint | get-woke/woke | WRITE | MIT | 298.7 KB | not yet graded | An agent detecting or replacing non-inclusive language in text/source queries this for its default rule set (terms, severities, alternatives) in YAML. |
| write-good | btford/write-good | WRITE | MIT | 172.1 KB | not yet graded | an agent tightening prose queries this for the rule behind a flag: passive voice, weasel words, wordiness, cliches, with the explanation |
| wysking--postgrad-study-engine | wysking/postgrad-study-engine | AI | MIT AND CC-BY-4.0 | 775.5 KB | not yet graded | an agent working on evidence-based study support queries this for idempotent learning records, diagnosis admission, memory fitting and budgeted prerequisite sch |
| wysking--quantitera | wysking/quantitera | API | MIT | 335.4 KB | not yet graded | an agent working on evidence-based economic research queries this for source/availability boundaries, prior expectations, knowledge supersession, scoped reviews |
| xerj-blogposts | xerj-org/xerj | WRITE | Apache-2.0 | 115.8 KB | not yet graded | an agent writing a XERJ blog post or docs page queries this for the house voice: how past posts open (a number first), argue (every claim traced to a run), and |
| xerj-columnar | ClickHouse/ClickHouse | DB | Apache-2.0 | 1.1 GB | not yet graded | an agent working on columnar layouts queries this for how clickhouse does codecs, layout and vectorised scans |
| xerj-search | quickwit-oss/tantivy, quickwit-oss/quickwit, meilisearch/meilisearch, valeriansaliou/sonic | DB | AGPL-3.0-only OR SSPL-1.0 OR Elastic-2.0+Apache-2.0+MIT+MIT AND BUSL-1.1+MPL-2.0 | 866.5 MB | not yet graded | an agent working on full-text search internals queries this for how lucene/tantivy/quickwit/meilisearch/sonic/elasticsearch do BM25, merge policy and the ES wir |
| xerj-storage | spacejam/sled, fjall-rs/fjall, cberner/redb | DB | Apache-2.0 OR MIT | 26.3 MB | not yet graded | an agent working on storage engines queries this for how sled/fjall/redb do WAL, crash recovery and compaction |
| xerj-vector | qdrant/qdrant, unum-cloud/usearch, instant-labs/instant-distance, nmslib/hnswlib | DB | Apache-2.0 | 42.0 MB | not yet graded | an agent working on vector search queries this for how qdrant/usearch/instant-distance/hnswlib do HNSW construction, quantisation and filtered kNN |
| xobotyi--cc-foundry-ai-helpers | xobotyi/cc-foundry | AI | MIT | 363.8 KB | not yet graded | an agent writing or maintaining Claude Code artifacts queries this for how to author prompts, Agent Skills, subagents and output styles: routing metadata that m |
| xobotyi--cc-foundry-ai-helpers-prompt-engineering | xobotyi/cc-foundry | AI | MIT | 63.8 KB | not yet graded | an agent writing instruction text for a model (a request, a system prompt, CLAUDE.md, a tool description, a brief for another agent) queries this for how to sta |
| xobotyi--cc-foundry-ai-helpers-skill-engineering | xobotyi/cc-foundry | AI | MIT | 91.6 KB | not yet graded | an agent authoring an Agent Skill queries this for the routing metadata that makes a skill activate, what belongs in SKILL.md against bundled references and scr |
| xobotyi--cc-foundry-backend | xobotyi/cc-foundry | OPS | MIT | 173.0 KB | not yet graded | an agent instrumenting a service queries this for observability rules written for agents: structured logging, Prometheus metric types, names and label cardinali |
| xobotyi--cc-foundry-cli | xobotyi/cc-foundry | OPS | MIT | 79.5 KB | not yet graded | an agent building a command-line tool or a shell script queries this for argument conventions, output streams, exit codes, configuration precedence and interact |
| xobotyi--cc-foundry-git-commit | xobotyi/cc-foundry | WRITE | MIT | 41.1 KB | not yet graded | an agent committing its work queries this for splitting a diff into atomic commits and ordering them, the quality gate before each commit, and commit message co |
| xobotyi--cc-foundry-golang | xobotyi/cc-foundry | API | MIT | 263.9 KB | not yet graded | an agent writing Go queries this for idioms, error handling, concurrency, testing, project layout and the toolchain features the module's go directive permits, |
| xobotyi--cc-foundry-grafana | xobotyi/cc-foundry | OPS | MIT | 528.6 KB | not yet graded | an agent building on Grafana queries this for dashboard JSON (Classic and V2), PromQL, VictoriaMetrics MetricsQL and VictoriaLogs LogsQL query writing, unified |
| xobotyi--cc-foundry-infrastructure | xobotyi/cc-foundry | OPS | MIT | 431.9 KB | not yet graded | an agent changing self-hosted infrastructure queries this for Ansible under the ansible-core line a project pins, Docker and Podman containers with Compose and |
| xobotyi--cc-foundry-javascript | xobotyi/cc-foundry | WEB | MIT | 470.6 KB | not yet graded | an agent writing JavaScript or TypeScript queries this for language semantics, the type system and tsconfig, the Node.js and Bun runtimes, and Vitest, each tied |
| xobotyi--cc-foundry-open-source | xobotyi/cc-foundry | WRITE | MIT | 95.8 KB | not yet graded | an agent contributing to a repository it does not maintain queries this for finding the project's AI-contribution rules before writing code, routing an issue an |
| xobotyi--cc-foundry-php | xobotyi/cc-foundry | API | MIT | 119.9 KB | not yet graded | an agent writing PHP queries this for modern PHP conventions, idioms and the type system, and for Pest and PHPUnit testing practice |
| xobotyi--cc-foundry-python | xobotyi/cc-foundry | API | MIT | 106.3 KB | not yet graded | an agent writing Python queries this for modern Python conventions, idioms, typing and the toolchain, and for pytest fixtures, parametrization and mocking |
| xobotyi--cc-foundry-rust | xobotyi/cc-foundry | API | MIT | 308.6 KB | not yet graded | an agent writing Rust queries this for ownership idioms, error handling, CPU parallelism with threads and rayon against async I/O with Tokio, the cargo, clippy |
| xobotyi--cc-foundry-the-blueprint | xobotyi/cc-foundry | AI | MIT | 191.8 KB | not yet graded | an agent planning a change before writing code queries this for the DRAFT pipeline (requirements discovery, objective codebase research, alignment on a directio |
| xobotyi--cc-foundry-the-blueprint-adr | xobotyi/cc-foundry | WRITE | MIT | 12.2 KB | not yet graded | an agent recording an architecture decision queries this for living ADRs that state the decision in force: when to write, rewrite or delete a record, the always |
| xobotyi--cc-foundry-the-coder | xobotyi/cc-foundry | API | MIT | 65.5 KB | not yet graded | an agent changing code in any language queries this for the language-agnostic workflow: discovery before editing, splitting a change into reviewable units, comm |
| xobotyi--cc-foundry-the-workflow-claude-md | xobotyi/cc-foundry | AI | MIT | 22.5 KB | not yet graded | an agent writing CLAUDE.md or .claude/rules files queries this for which layer loads when, what content belongs in each, and why a stated rule gets ignored, cit |
| xobotyi--cc-foundry-the-workflow-debrief | xobotyi/cc-foundry | AI | MIT | 212.0 KB | not yet graded | an agent improving its own instructions queries this for how to harvest user corrections, retractions, tool errors and repeated procedures from Claude Code sess |
| xobotyi--cc-foundry-the-workflow-handoff | xobotyi/cc-foundry | AI | MIT | 7.2 KB | not yet graded | an agent about to cross a context boundary (a session restart, a delegation, a compaction) queries this for how to triage a conversation into a standalone trans |
| xobotyi--cc-foundry-the-writer | xobotyi/cc-foundry | WRITE | MIT | 37.7 KB | not yet graded | an agent drafting prose people read queries this for the AI-writing tells, the mechanisms that produce them, the thresholds that mark a passage, and edits that |
| xz | tukaani-project/xz | NET | 0BSD | 6.7 MB | not yet graded | NET |
| zalando-restful-api-guidelines | xerj-org/corpus-zalando-guidelines | API | CC-BY-4.0 | 327.7 KB | G7 1.0/5 | an agent reviewing an API design queries this for the rule, its intent and propulated exceptions from a fleet of thousands |
| zlib-ng | zlib-ng/zlib-ng | NET | Zlib | 8.7 MB | not yet graded | an agent working on deflate queries this for the modernised algorithms and SIMD paths |
| zstd | facebook/zstd | NET | mixed: BSD-3 (lib) / GPL-2.0 (cli) | 8.1 MB | not yet graded | an agent tuning compression queries this for frame/block formats, window and level internals |
G7 is 5 pre-registered queries, graded on the top 5 results.
Contributed corpora · 12,642
Repositories their authors agreed to share, listed on each domain page. SEC 335 · DB 459 · NET 451 · CRYPTO 201 · WEB 2,092 · API 4,766 · OPS 658 · LAW 49 · HEALTH 77 · AI 3,364 · WRITE 190
For agents
llms.txt (the catalogue and how to ask it) · llms-repos.txt (every pinned repository) · llms-contributors.txt (every contributor)