auto-conclusion

Generate structured development summaries from conversations and Git branch or commit diffs.

Updated Aug 1, 2026
One-click install
npx skills add https://github.com/lov-b/Skills --skill auto-conclusion-lov-b
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: auto-conclusion
Source: https://github.com/lov-b/Skills/tree/main/skills/auto-conclusion
Command: npx skills add https://github.com/lov-b/Skills --skill auto-conclusion-lov-b

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? After finishing a bug fix or feature, developers often lose the reasoning, failed attempts, and validation details scattered across long AI conversations and multiple Git branches. This Skill consolidates that context into a complete, well-structured Markdown summary document. ## Core Features & Use Cases - Multi-source summarization: Combines conversation history with Git branch or commit diffs (single or multi-repository), writing the development narrative first and keeping branch/commit details as appendices. - Coverage auditing: Scans both user questions and agent answers to extract derived questions, trade-offs, risks, and boundaries, then deduplicates them via a coverage matrix before writing. - Business Q&A and knowledge archiving: Archives project-specific business questions into matched existing documents (with classification protection for legacy content) and routes reusable technical knowledge into categorized knowledge files. - Optional Git integration: Detects the documentation repository after writing and can commit and push only the generated summary file. - Use Case: After finishing a cross-repo feature, say "总结这些分支并带上对话" to get a complete summary covering requirements, solutions, pitfalls, test notes, and commit messages, saved to your docs folder and optionally pushed. ## Quick Start Ask the agent to summarize the bug or feature just completed in this conversation, optionally naming specific branches or commits to include their diffs.

Frequently Asked Questions about auto-conclusion

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I summarize a bug fix or feature from an AI conversation?

Say phrases like "总结刚刚完成的需求" or "总结这个对话中解决的bug" to trigger the skill. It retrieves relevant conversation rounds, filters out unrelated tangents, and writes a structured Markdown document covering requirements, solutions, problems, and test notes.

How to generate a summary from Git branches or commits?

Name the branches or commit hashes and their repository paths, and the skill runs git diff against a baseline (default main or master) for each project. It then merges the diffs with related conversation context, keeping change details in appendices.

Can it update an existing summary document with new conversation content?

Yes, document supplement mode reads the existing file, extracts its requirement anchors, filters the current conversation by those anchors, and writes additions back to the original file. New sections are labeled with the supplement date.

Does it automatically commit and push the summary to Git?

After writing the file, it detects whether the target directory is inside a Git repository and asks whether to commit and push. Only the generated document is staged, and push failures are reported without rolling back the file.

Where are the summary documents saved by default?

Save location resolves by priority: a user-specified path, a configured defaultSaveDir in the user-level config.json, or the system Downloads folder. Knowledge documents go into a separate knowledge subdirectory.

What is the difference between business Q&A archiving and knowledge summarization?

Business Q&A archiving stores project-specific decisions and logic into matched existing documents, while knowledge summarization extracts reusable, project-independent technical concepts into categorized knowledge files. The skill routes each candidate item to the correct destination.