little-canary
Detects prompt injection by its effect on a sacrificial canary model, not just pattern matching: untrusted input hits a powerless model first, a behavioral check reads the residue, and it returns block, flag, or pass before your primary model acts. Inbound preflight sensor, not a guarantee
What XERJ indexed
Snapshot
The whole repository at commit 6038b1011b (2026-09-30). main is 45 commits ahead.
Selected work
“Add FastAPI and LangChain integration examples”. 3 files, 517 lines in examples, tests.
Licence
The repository
LICENSE file checked at the pinned commit; the author agreed to inclusion.
The XERJ corpus
The author agreed by email on 2026-10-24. Same terms as the repository. Answers name the repo, the commit and the file:line.
selected for the XERJ Corpus Hublicence clearedsuggested use: adapt with attribution
ai-safetydetectionprompt-injectionllm-securityagent-securityai-agentscanarydefense-in-depth
Example questions
- How does little-canary implement ai-safety?
- Where does little-canary handle “Add FastAPI and LangChain integration examples”?
- Show the detection code in little-canary.
Inside the indexed corpus
- Languages by size: Python 96.5%, JavaScript 2.3%, HTML 1.2%.
- On GitHub since 2026-02-23, last push 2026-10-26, 266 commits on main.
- The pin is 45 commits behind main; the corpus answers from the pinned code, not the newest.
Licence on GitHub and in the corpus
GitHub reported Apache-2.0 for hermes-labs-ai/little-canary on 2026-10-29, and the LICENSE file at commit 6038b1011b was reviewed as Apache-2.0. They match, so answers from this corpus carry the same terms as the live repository.
Contributors
Rashmeet Chhabra (@rashmeetchhabra12) has 1 selected repository on the hub.
Using XERJ with little-canary
- RAG without a vector database: one hybrid index The hub files it under AI, prompts and agents.
- Index a reference library for a coding agent
- ripgrep vs a code index for coding agents: 30 tasks
- Best MCP servers for local code search in 2026, compared
The prompt
Paste it into your AI agent. It clones hermes-labs-ai/little-canary at the pinned commit and indexes it on your machine.
Search the GitHub repository hermes-labs-ai/little-canary with XERJ (https://hub.xerj.org/corpus/hermes-labs-ai--little-canary). Run these in a shell, in order, and check each result before the next: 1. xerj --help (success: it prints usage). If xerj is missing: curl -fsSL https://xerj.org/get | sh 2. git clone https://github.com/hermes-labs-ai/little-canary little-canary && git -C little-canary checkout 6038b1011bef8b9fbe065b714df45c6deff5fc82 3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data 4. xerj autoindex ./little-canary --url http://127.0.0.1:9200 --prefix little-canary --progress plain 5. xerj autoindex map --url http://127.0.0.1:9200 Then answer my question from the indexed files and cite the ax_path of every hit. Query syntax: https://xerj.org/llms.txt Licence: Apache-2.0. Keep the copyright notice and credit the authors.
Related
Rashmeet ChhabraAI, prompts and agents corporaagnosplunk-security-contentgaraksandeepmothukuri/PromptSentinelsaintparish4/openscriptZwony/RedAgent
Stats: GitHub API, fetched 2026-10-29. Selected work and consent: the XERJ registry.