$ loading_
Detects code that deviates from repo patterns learned from git history.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "argot" MCP server from askskill: Run: claude mcp add 'io-github-get-tmonier-argot' -- npx -y @tmonier/argot
Before reviewing a commit or PR, developers can use it to find code that does not match historical repository patterns and narrow the manual review scope. It is useful for prioritizing changes that look unlike the project’s usual style.
When a team wants a quick local check for code that diverges from existing implementation conventions, this tool can help. It runs locally using statistical methods and does not rely on an LLM.
It learns existing code patterns from a repository’s Git history and flags code that does not match those patterns. The description emphasizes that it runs locally using statistical methods and does not use an LLM.
The provided description says it runs locally and uses statistics rather than an LLM. For any other network-related dependencies, see the source repository.
Based on the description, the main prerequisite is a repository with Git history, since it relies on historical patterns to detect anomalies. For installation and runtime requirements, see the source repository.
Run exploratory browser tests and create reproducible bug reports for coding agents.
Ask natural-language questions over codebases with citations and impact analysis.
Analyze codebases and generate AGENTS.md files for AI coding agents.
Search code efficiently and apply safe atomic edits across multiple files.
Enable AI agents and IDEs to search codebases with hybrid retrieval and precise line-level citations.
Help AI analyze local code context, assess impact, and judge merge readiness.