Agent skill

auto-improvement

(ePost) Use when reviewing session performance, analyzing improvement trends, or understanding the metrics-capture-detection pipeline

View SKILL.md on GitHub Repository
Stars 163
Forks 31

Install this agent skill to your Project

npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/auto-improvement

Metadata

Additional technical details for this skill

keywords
metrics improvement lessons trends detection rework session
triggers 
[
    "improvement report",
    "session metrics",
    "detect patterns",
    "rework analysis"
]
platforms 
[
    "all"
]
agent affinity 
[
    "epost-project-manager",
    "epost-code-reviewer",
    "epost-debugger"
]

SKILL.md

Auto-Improvement Skill

Purpose

Documents the automated feedback loop: metrics collection, lesson capture, and pattern detection across sessions.

Architecture

SessionStart hook          Stop hooks (sequential)
      │                          │
  write marker ──→   session-metrics.cjs ──→ lesson-capture.cjs
  current-session.json       │                      │
                        append JSONL           evaluate thresholds
                        sessions.jsonl              │
                                             significant? ──→ prompt knowledge-capture
                                                  │
                                               /review-improvements
                                             (on-demand detection)

Data Flow

Storage Tiers

Tier Location Tracked Retention
Raw metrics .epost-data/improvements/sessions.jsonl No (gitignored) Auto-rotate at 1000 lines
Session marker .epost-data/improvements/current-session.json No Deleted on Stop
Detection report .epost-data/improvements/latest-report.json No Overwritten per run
Significant learnings docs/ Yes (git tracked) Permanent

JSONL Schema

Each line in sessions.jsonl is a JSON object:

json
{
  "sessionId": "string",
  "timestamp": "ISO-8601",
  "duration_ms": "number|null",
  "branch": "string",
  "git": { "filesChanged": 0, "insertions": 0, "deletions": 0 },
  "tasks": { "total": 0, "completed": 0, "failed": 0 },
  "errors": { "count": 0, "types": ["build|test|lint|runtime"] },
  "rework": { "fixIterations": 0, "verificationFailures": 0 },
  "skills": { "discovered": [], "loaded": [], "unused": [] },
  "knowledge": { "retrieved": 0, "captured": 0, "staleHits": 0 },
  "routing": { "intent": "string", "command": "string", "platform": "string" }
}

Full schema: packages/core/assets/improvements-schema.json

Significance Thresholds

Lesson capture triggers automatically when:

Trigger Threshold Capture Type
Errors fixed errors.count > 0 FINDING
Rework detected fixIterations >= 2 PATTERN
Verification failure verificationFailures >= 1 CONV
New skill first-seen Not in previous 5 sessions NOTE

Detection Engine

Analyzes accumulated metrics from .epost-data/improvements/sessions.jsonl:

Detection Method Recommendation
Repeat errors Same error type 3+ times in 7 days Create FINDING or fix root cause
Skill gaps Platform detected but no platform-skill loaded Check skill wiring
Stale knowledge docs/ entries >90 days with no references Review or archive
Rework patterns Avg fixIterations >2 over recent sessions Review verification process
Unused skills Loaded but unreferenced 10+ sessions Optimize or remove

Run via command: /review --improvements

Hooks

Hook Event Purpose
session-init.cjs SessionStart Writes current-session.json marker
session-metrics.cjs Stop Collects metrics, appends to JSONL
lesson-capture.cjs Stop (prompt) Evaluates significance, prompts capture

Related Skills

  • data-store — .epost-data/ convention for persistent agent data
  • knowledge-capture — Post-task capture workflow (triggered by lesson-capture hook)
  • knowledge-retrieval — Search existing knowledge before acting
  • knowledge-retrieval — Knowledge system structure and conventions
  • verification-before-completion — Verification failures tracked for rework detection
  • debug — Error metrics auto-captured by session-metrics hook
  • code-review — Convention violations auto-detected across sessions

Recommended Agent Skills

Expand your agent's capabilities with these related and highly-rated skills.

majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-spec

Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-testing

Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.

163 31
Explore
majiayu000/claude-skill-registry

agent-ops-state

Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.

163 31
Explore

Didn't find tool you were looking for?

Be as detailed as possible for better results