菜单
Capture a correction or lesson as a persistent learning rule with category, mistake, and correction. Stores, categorises, and retrieves rules for future sessions. Use after mistakes or when the user says "remember this", "don't forget", "note this", or "learn from this".
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
SkillOpt-flavored offline training loop for any SKILL.md. Treats accumulated learn-rule corrections as training trajectories, proposes bounded patches via an optimizer LLM, gates each candidate against a held-out validation set built from the user's own past corrections, and ships only candidates that demonstrably improve the score. Inspired by Microsoft SkillOpt's ReflACT pipeline (rollout → reflect → aggregate → select → update → evaluate) adapted to pro-workflow's SQLite store. Use when a skill has accumulated 8+ learn-rule rows and the user wants the skill itself to get better, not just longer.
Auto-configure quality gates, hooks, and settings for a new project. Detects project type and sets up appropriate tooling. Use when onboarding a new codebase.
Analyze permission denial patterns and generate optimized alwaysAllow and alwaysDeny rules. Use when permission prompts are slowing you down or after sessions with many denials.
Complete AI coding workflow system. Orchestration patterns, 18 hook events, 5 agents, cross-agent support, reference guides, and searchable learnings. Works with Claude Code, Cursor, and 32+ agents.
Prevent destructive operations using Claude Code hooks. Three modes — cautious (warn on dangerous commands), lockdown (restrict edits to one directory), and clear (remove restrictions). Uses PreToolUse matchers for Bash, Edit, and Write.
Render a self-contained HTML viewer for a pro-workflow wiki. Pages, sources, claims, seed queue, page-link graph and full-text search all in one file. No external dependencies, no JS framework, S3-uploadable. Use when the user wants to browse a wiki visually, share its current state with someone, audit research progress, or hand off a knowledge base. Inspired by Thariq Shihipar's "Unreasonable Effectiveness of HTML" — favors information density and shareability over markdown-only outputs.
| name | learn-rule |
| description | Capture a correction or lesson as a persistent learning rule with category, mistake, and correction. Stores, categorises, and retrieves rules for future sessions. Use after mistakes or when the user says "remember this", "don't forget", "note this", or "learn from this". |
Capture a lesson from the current session into permanent memory.
Use when the user says "remember this", "add to rules", "don't do that again", or after a mistake is identified.
[LEARN] Category: One-line rule
Mistake: What went wrong
Correction: How it was fixed
Append Wiki: <slug> to bind the rule to a single pro-workflow wiki. The rule loads only when that wiki is in scope, avoiding cross-project pollution:
[LEARN] Editing: Cite a sources.md row before adding any wiki claim.
Wiki: agent-memory
The capture hook auto-detects Wiki: <slug> and links the learning to that wiki via learnings_wiki.
| Category | Examples |
|---|---|
| Navigation | File paths, finding code, wrong file edited |
| Editing | Code changes, patterns, wrong approach |
| Testing | Test approaches, coverage gaps, flaky tests |
| Git | Commits, branches, merge issues |
| Quality | Lint, types, style violations |
| Context | When to clarify, missing requirements |
| Architecture | Design decisions, wrong abstractions |
| Performance | Optimization, O(n^2) loops, memory |
Recent mistake: Edited wrong utils.ts file
[LEARN] Navigation: Confirm full path when multiple files share a name.
Add to LEARNED section? (y/n)
[LEARN] rule with category