بنقرة واحدة
experience-evolution
Project knowledge accumulation system - learn from practice, avoid repeating mistakes
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
القائمة
Project knowledge accumulation system - learn from practice, avoid repeating mistakes
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
File-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use for multi-step tasks requiring >5 tool calls.
Session wrap-up - update handoff + commit + auto-record experience
Systematic debugging - four-phase process, find root cause before fixing
Verification before completion - must run verification commands before claiming done. Evidence before claims.
| name | experience-evolution |
| description | Project knowledge accumulation system - learn from practice, avoid repeating mistakes |
| version | 1.0.0 |
| mode | observe-only |
Current mode: Observe-Only Won't modify any files, only provides learning summary at session end.
Automatically remember your successes and failures:
Command Execution (Bash)
File Modifications (Write/Edit)
Working Directory
Session Learning Summary
Project: your-project
Duration: 45 minutes
Commands: 15 (13 success, 2 failed)
Reusable Patterns Found:
1. API Retry Mechanism (High Value)
Scenario: API QUOTA_EXCEEDED
Solution: Exponential backoff retry (1s, 2s, 3s)
Effect: 95% of rate limit errors auto-recover
Reuse potential: All external API calls
2. TypeScript Type Fix (Medium Value)
Scenario: Union type narrowing issues
Solution: Type guard functions
Effect: Eliminated 12 type errors
Reuse potential: Medium (specific to current structure)
Save these learnings?
[y] Yes, save to knowledge base (confirm each)
[n] No, skip this time
[v] View details
Phase 1 (Current): Observe Only — No file writes, session-end summaries
Phase 2: Write Logs — Record to learned/ directory, manual confirmation
Phase 3: Semi-auto — Low-risk auto-save, high-risk needs confirmation
Phase 4: Cross-project — Experience from project A auto-suggests in project B
Will never:
Will only: