by infinri · Agent Tool · ★ 103
Writ A Claude Code harness that gives every coding session two helpers: a fast librarian that picks the rules that fit the current task, and a process keeper that blocks risky writes until you have approved a plan and tests. At the live 276-rule production corpus (post Phase 1-5 public-rulebook expansion), the librarian returns ranked results in 0.590 ms at the 95th percentile. At the 10,000-rule synthetic scale, it still holds at 0.557 ms while reducing context tokens by 726 times versus loading the whole rulebook every turn.
| Stars | 103 |
| Forks | 8 |
| Language | Python |
| Category | Agent Tool |
| License | MIT |
| Quality Score | 68.2648362163134/100 |
| Last Updated | 2026-05-22 |
| Created | 2026-03-16 |
| Platforms | claude-code, python |
| Est. Tokens | ~151k |
These tools work well together with Writ for enhanced workflows:
Explore other popular agent tool tools:
Writ is 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 rule. It is categorized as a Agent Tool with 103 GitHub stars.
Writ is primarily written in Python. It covers topics such as agent-tools, ai-coding, bm25.
You can find installation instructions and usage details in the Writ GitHub repository at github.com/infinri/Writ. The project has 103 stars and 8 forks, indicating an active community.
Writ is released under the MIT license, making it free to use and modify according to the license terms.