borghei/Claude-Skills / engineering/self-improving-agent
self-improving-agent
VerifiedCurates agent memory and turns repeated lessons into reviewed rules: memory health checks, pattern extraction, feedback and regression analysis. Use when managing auto-memory or building agents that learn from their own sessions.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 10.0 KB |
| assets/memory_index/MEMORY.md | markdown | 316 B |
| assets/memory_index/feedback_migrations.md | markdown | 203 B |
| assets/memory_index/feedback_pnpm.md | markdown | 255 B |
| assets/sample_feedback.jsonl | other | 2.2 KB |
| assets/sample_metrics.json | config | 3.6 KB |
| references/feedback-loop-patterns.md | markdown | 9.2 KB |
| references/memory-curation-guide.md | markdown | 4.6 KB |
| references/meta-learning-architectures.md | markdown | 2.4 KB |
| references/self-improvement-methodology.md | markdown | 4.3 KB |
| scripts/feedback_analyzer.pyexecutable | script | 18.5 KB |
| scripts/memory_health_checker.py | script | 13.1 KB |
| scripts/pattern_extractor.py | script | 10.8 KB |
| scripts/regression_detector.pyexecutable | script | 17.0 KB |
| scripts/rule_manager.pyexecutable | script | 12.0 KB |
| scripts/rule_promoter.py | script | 12.0 KB |
| skills/extract.md | markdown | 2.1 KB |
| skills/promote.md | markdown | 2.8 KB |
| skills/remember.md | markdown | 2.1 KB |
| skills/review.md | markdown | 2.3 KB |
| skills/status.md | markdown | 2.5 KB |
skills/status.md
View on GitHub# Sub-Skill: Status
**Parent:** self-improving-agent
**Trigger:** "memory status", "learning progress", "how am I improving", "show improvement metrics"
## Purpose
Display the current state of the self-improvement system: memory size, rule counts, learning velocity, and improvement trends. Provides a dashboard view of agent learning progress.
## Workflow
### Step 1: Gather Metrics
Collect data from all improvement system components:
**Memory metrics:**
- Total entries in MEMORY.md
- Entries per topic file
- Line counts vs limits
- Oldest and newest entries
**Rule metrics:**
- Total promoted rules
- Rules by category
- Rules by age (fresh / aging / stale)
- Rules with/without "why" annotations
**Learning velocity:**
- Entries added in last 7 days
- Entries promoted in last 30 days
- Entries pruned in last 30 days
- Net growth rate
### Step 2: Compute Improvement Score
```
Improvement Score = (
promoted_rules_30d * 0.3 +
(1 - stale_entry_ratio) * 0.2 +
first_attempt_success_delta * 0.3 +
memory_health_score * 0.2
)
```
Where:
- `promoted_rules_30d`: Rules promoted in last 30 days (normalized 0-1)
- `stale_entry_ratio`: Fraction of entries classified as stale
- `first_attempt_success_delta`: Change in first-attempt success rate
- `memory_health_score`: 1.0 if under limits, decreasing with violations
### Step 3: Determine Maturity Level
Map to the improvement maturity model:
| Level | Name | Indicator |
|-------|------|-----------|
| 0 | Stateless | No MEMORY.md or empty |
| 1 | Recording | Entries exist, no promotions |
| 2 | Curating | Regular reviews, entries classified |
| 3 | Promoting | Active rule promotion pipeline |
| 4 | Extracting | Skills being extracted from patterns |
| 5 | Meta-Learning | Capture strategy adapting based on value |
### Step 4: Display Dashboard
```
Self-Improvement Status
========================
Maturity Level: 3 (Promoting)
Improvement Score: 0.72
Memory: 47 entries (187/200 lines)
Rules: 12 promoted (3 this month)
Velocity: +8 entries, -3 pruned, +3 promoted (30d)
Health: GOOD (no constraint violations)
Trend: IMPROVING (first-attempt success +5% over 30d)
```
## Inputs
| Input | Required | Description |
|-------|----------|-------------|
| Memory path | No | Defaults to ./MEMORY.md |
| Feedback log | No | Path to feedback data for trend analysis |
## Outputs
- Dashboard with current metrics
- Maturity level classification
- Improvement score with breakdown
- Trend direction (improving / stable / degrading)
- Recommended next actions
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |