What problem does it solve? Configuring Claude Code permissions manually is tedious and error-prone: you must know which read-only bash commands are safe to allow, which package manager the project uses, and which documentation domains to whitelist. This Skill automates that audit by scanning the repository and producing a ready-to-use settings.json. ## Core Features & Use Cases - Tech Stack Detection: Scans for indicator files (pyproject.toml, package.json, go.mod, Cargo.toml, lock files) to identify languages, frameworks, package managers, and monorepo tools. - Permission Generation: Builds an allow list of read-only bash commands (git, gh CLI, package manager queries) matched strictly to the detected stack, excluding state-modifying commands. - MCP & WebFetch Suggestions: Recommends Sentry or Linear MCP server configs for .mcp.json and framework-specific documentation domains for WebFetch. - Use Case: When onboarding a new Python/Django project, run the audit to instantly get a settings.json allowing poetry, git, and gh read commands plus docs.djangoproject.com fetches, without hand-writing any JSON. ## Quick Start Analyze this repository and generate a recommended Claude Code settings.json with read-only permissions for the detected tech stack.
