infinri/Writ — Claude Code harness for AI coding agents.

Description and use

**Project Summary**

infinri/Writ is an open-source project that provides a Claude Code harness for AI coding agents. It offers a hybrid-RAG librarian that indexes knowledge bases, enabling efficient search and retrieval of relevant information. This framework helps make internal knowledge searchable and usable by assistants, improving the accuracy and reliability of AI-generated responses.

**Value Proposition**

The value proposition of infinri/Writ lies in its ability to index knowledge bases, making it easier to search and retrieve relevant information. This can be particularly useful in scenarios where AI assistants need to access and utilize internal knowledge to provide accurate responses. By using infinri/Writ, organizations can improve the quality of their AI-generated responses and reduce the time spent on manual knowledge retrieval.

**Practical Adoption Path**

To adopt infinri/Writ, organizations can follow these steps:

1. Evaluate the project's implementation signals, such as API/SDK/CLI, language metadata, or focused topics, to determine its feasibility for their specific use case.
2. Assess the project's production readiness, considering factors such as recent activity, adoption, and ecosystem signals.
3. Review the project's quality signals, including GitHub stars, forks, and updates, to gauge its maintainability and community support.
4. Conduct a final review of

Problem

infinri/Writ helps make internal knowledge searchable and usable by assistants.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• index knowledge bases
• improve search over documents
• ground assistant answers

Evidence

Claude Code harness for AI coding agents. Hybrid-RAG librarian picks relevant rules in 0.59 ms p95 (BM25 + vector + graph traversal); process keeper blocks writes until plan + tests approved. 276 rules / 30 mandatory across 12 domains, 6 cross-language analyzers. Neo4j + Tantivy + hnswlib + ONNX.

Source