Analyze coding session histories to detect convention violations and propose configuration updates.
npx skills add https://github.com/bbvch-ai/aihub-core --skill reflect-bbvch-ai
Please help me install this Agent Skill. Skill: reflect Source: https://github.com/bbvch-ai/aihub-core/tree/main/.claude/skills/reflect Command: npx skills add https://github.com/bbvch-ai/aihub-core --skill reflect-bbvch-ai
This Skill solves the problem of repeating the same coding mistakes across sessions by identifying where the AI assistant violated project conventions, missed established patterns, or required user steering, eliminating wasted time on recurring issues.
Use the reflect skill after a coding session to get a prioritized list of actionable improvements for your project's documentation and AI configuration.
High-intent search queries and answers about installing and using this skill.
To stop an AI assistant from repeatedly violating project coding conventions, you can analyze session histories to detect breaches and generate permanent configuration proposals for your CLAUDE.md files, skills, or hooks to prevent future recurrences.
You can analyze a coding session history by scanning it for instances where the AI violated architectural patterns or required user steering, then categorizing those issues to propose targeted systemic fixes to your project's AI configuration.
The best way to update CLAUDE.md files is to run a retrospective analysis on your session history to identify convention violations, which generates categorized, actionable proposals for updating your documentation to prevent recurring mistakes.
Yes, you can detect Pydantic usage and type hint standard violations by scanning past session histories against your project's core coding conventions to identify where the AI assistant failed to follow established patterns.
Yes, this convention violation detection applies specifically to software development workflows using the Swiss AI-Hub core codebase, covering violations of type hint standards, Pydantic usage, scope boundaries, and tooling conventions.
You should run a retrospective on your AI coding sessions immediately after the session ends, allowing you to capture fresh context on convention violations and generate prioritized actionable improvements for your project's documentation and AI configuration.
nq-rdlCoordinate Jules AI coding sessions to dispatch tasks and collect results.
anam-orgAnalyze sessions to propose CLAUDE.md, hook, and skill improvements.
bbvch-aiGenerate FastAPI REST controllers with DTOs, RBAC, and entrypoint registration.
zhaozengqing4364-bitAnalyze conversation logs to identify usage patterns, errors, and improvement opportunities.
