flai
FLAI is a self-hosted, privacy-first AI platform. Your all-in-one local assistant: smart chat, web search, voice, media generation, and document analysis. Offers long-term memory, Model Hub, and OpenAI API. Fully optimized for GPU and CPU environments
What XERJ indexed
Selected work
“render message media, previews and tasks, and stop the mic…”. 4 files, 291 lines in demo.
Licence
The repository
LICENSE file checked at the pinned commit; the author agreed to inclusion.
The XERJ corpus
The author agreed by email on 2026-10-27. 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
dockerdocker-composepiper-ttspython3qdrantrediswhisperroom-snapshot-api
Example questions
- How does flai implement docker?
- What does demo do in flai?
- Show the docker-compose code in flai.
Inside the indexed corpus
- 537 files, 33.4 MB in total, about 63.7 KB per file.
- The selected commit changed 4 of those files (0.7%).
- Languages by size: Python 82.5%, JavaScript 10.2%, CSS 2.9%, Shell 2.4%, HTML 1.9%.
- On GitHub since 2026-03-10, last push 2026-10-29, 1,754 commits on main.
Licence on GitHub and in the corpus
GitHub reported MIT for barval/flai on 2026-10-31, but the LICENSE file at commit 2af99cb568 was reviewed as NOASSERTION. The corpus is the pinned commit, so answers from it follow NOASSERTION.
Contributors
Valery Barsukov (@barval) has 2 selected repositories on the hub.
Their other repository pages: barval/room-snapshot-api.
Using XERJ with flai
- Find every place a config key is set in a repo flai is tagged docker-compose on GitHub.
- Search an OpenAPI spec so an agent can find endpoints The hub files it under API design and specification.
- Best grep alternative for AI coding agents (2026)
- Ask questions about a GitHub repo without cloning it
The prompt
Paste it into your AI agent. It clones barval/flai at the pinned commit and indexes it on your machine.
Search the GitHub repository barval/flai with XERJ (https://hub.xerj.org/corpus/barval--flai). 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/barval/flai flai && git -C flai checkout 2af99cb568adfe13433dcb170d0ff95218ec63b1 3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data 4. xerj autoindex ./flai --url http://127.0.0.1:9200 --prefix flai --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: check the LICENSE file. Keep the copyright notice and credit the authors.
Related
Valery BarsukovAPI design and specification corporadunorth595--ai-agent-chatann-benchmarksxobotyi--cc-foundry-infrastructureDigitalJacob/booking_botChaitanyMishra/invoice-receipt-parserArshia-Moghbel/Telegram-VPN-Seller-Bot
Stats: GitHub API, fetched 2026-10-31. Snapshot: the git tree at the pinned commit. Selected work and consent: the XERJ registry.