القائمة
Cross-project learning engine with automatic failure capture and context-aware memory compression
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
Manage projects on Nulab Backlog — issues, wikis, milestones, Git, sprints for Japan's top PM tool
Automate Japan's
Unified Gmail, Calendar, Drive, Sheets, and Docs management from Claude Code
Task and project management on Jooto — create tasks, manage boards, automate workflows via PR TIMES' tool
Attendance and time tracking automation — daily/monthly data, payroll sync via Japan's leading attendance system
Build, query, and automate Cybozu kintone apps — Japan's leading no-code business platform
| name | Self-Learning Agent |
| description | Cross-project learning engine with automatic failure capture and context-aware memory compression |
| author | hanabi-jpn |
| version | 1.0.0 |
| tags | ["ai","learning","memory","cross-project","self-improvement"] |
╭──────────────────────────────────────────╮
│ │
│ 🧠 S E L F - L E A R N I N G 🧠 │
│ A G E N T │
│ │
│ ┌─────────┐ │
│ │ 📚 📚 │ "I remember │
│ │ 💡 │ everything." │
│ │ ╰───╯ │ │
│ └────┬────┘ │
│ ┌───────┼───────┐ │
│ ▼ ▼ ▼ │
│ [PJ-A] [PJ-B] [PJ-C] │
│ └───────┼───────┘ │
│ 🌐 GLOBAL │
╰──────────────────────────────────────────╯
🧠 Cross-Project ⚡ Auto-Capture 📊 Analytics 🗜 Compressed v1.0.0
Cross-project learning engine with automatic failure capture, intelligent knowledge promotion, and context-aware memory compression.
Author: hanabi-jpn | Version: 1.0.0 | License: MIT
Tags: ai learning memory cross-project self-improvement
Self-Learning Agent captures errors, corrections, and patterns across ALL your projects — not just one. It automatically detects failures, logs learnings, promotes cross-project knowledge, and compresses context to prevent memory bloat.
┌─────────────────────────────────────────────────┐
│ LEARNING PIPELINE │
│ │
│ ┌──────────┐ ┌───────────┐ ┌──────────┐ │
│ │ CAPTURE │───▶│ ANALYZE │───▶│ STORE │ │
│ │ Auto/Man │ │ Categorize│ │ Project │ │
│ └──────────┘ └───────────┘ └──────────┘ │
│ │ │
│ ▼ │
│ ┌──────────┐ ┌───────────┐ ┌──────────┐ │
│ │ APPLY │◀───│ PROMOTE │◀───│ SCORE │ │
│ │ Context │ │ Proj→Glob │ │ Relevance│ │
│ └──────────┘ └───────────┘ └──────────┘ │
│ │ │
│ ▼ │
│ ┌──────────────────────────────────────────┐ │
│ │ COMPRESS — Keep context under budget │ │
│ └──────────────────────────────────────────┘ │
└─────────────────────────────────────────────────┘
You are an agent equipped with Self-Learning Agent, a cross-project learning system. Follow these rules:
After EVERY tool execution that results in an error:
Capture the error context:
Categorize the error:
syntax — Code syntax errorsruntime — Runtime failures, crashesconfig — Configuration issues, missing env varsnetwork — API failures, timeouts, DNS errorspermission — File/directory permission issuesdependency — Missing packages, version conflictslogic — Wrong output, unexpected behaviorSearch for similar past errors:
<project>/.self-learning/learnings.jsonl~/.openclaw/self-learning/global/errors.jsonlLog the learning:
{
"id": "learn-{timestamp}",
"type": "error",
"category": "runtime",
"context": "Running pytest on auth module",
"error": "ImportError: cannot import name 'jwt' from 'jose'",
"fix": "Install python-jose[cryptography] instead of python-jose",
"project": "web-api",
"frequency": 1,
"impact": 0.8,
"created": "2026-03-01T10:00:00Z",
"status": "active"
}
Knowledge lives at two levels:
Global (~/.openclaw/self-learning/global/):
errors.jsonl — Errors seen across multiple projectspatterns.jsonl — Workflow patterns that work everywherepreferences.json — User preferences (coding style, tools, etc.)index.json — Fast lookup index by category and keywordProject (<project>/.self-learning/):
learnings.jsonl — Project-specific learningsoverrides.json — Project-specific settings that override global defaultsKnowledge Flow:
Each learning has a promotion score:
score = frequency × impact × recency_weight
where:
frequency = times this learning was useful (1-10, capped)
impact = severity of the problem it solves (0.0-1.0)
recency = 1.0 if used today, decays 0.1 per week of non-use
Promotion rules:
.self-learning/archive/)CRITICAL: Never load more than 2000 tokens of learning context per session.
Compression strategies:
learn — Manually log a learning:
> learn
🧠 What did you learn?
User: Always use --platform linux/amd64 with Docker build on M1 Mac
📝 Learning captured:
Category: config
Context: Docker builds fail on Apple Silicon without explicit platform
Fix: Add --platform linux/amd64 to docker build commands
Impact: 0.7
Project: web-api
✅ Stored in web-api/.self-learning/learnings.jsonl
🔍 Similar global learning found — merged (frequency: 3 → 4)
learn --auto — Toggle automatic learning mode:
.self-learning/config.json> learn --auto
🔄 Auto-learning mode: ON → OFF
Previous state: Capturing errors and corrections automatically
New state: Manual logging only (use `learn` to log manually)
Config saved to .self-learning/config.json
To re-enable: run `learn --auto` again
learn recall <topic> — Search knowledge base:
> learn recall docker
🔍 Searching knowledge base for "docker"...
Results (5 matches):
[1] ⭐ Score: 9.2 | Global | config
"Always set --platform linux/amd64 on Apple Silicon"
Fix: docker build --platform linux/amd64 -t myapp .
Used: 7 times across 4 projects
[2] Score: 7.5 | Global | runtime
"Docker Compose v2 uses 'docker compose' (no hyphen)"
Fix: Replace 'docker-compose' with 'docker compose'
Used: 4 times across 3 projects
[3] Score: 6.1 | Project: web-api | config
"Set DOCKER_BUILDKIT=1 for faster multi-stage builds"
Fix: export DOCKER_BUILDKIT=1 before build commands
Used: 2 times
[4] Score: 4.3 | Project: ml-pipeline | runtime
"Docker volume mounts fail silently with wrong paths"
Fix: Use absolute paths and verify with docker inspect
Used: 1 time
[5] Score: 2.1 | Project: web-api | dependency (archived)
"Pin base image versions in Dockerfile"
Fix: Use python:3.11-slim instead of python:latest
Used: 1 time (last used: 45 days ago)
learn stats — Show learning analytics:
╔═══════════════════════════════════════════╗
║ Self-Learning Agent Stats ║
╠═══════════════════════════════════════════╣
║ Total Learnings: 142 (37 global) ║
║ Categories: ║
║ errors: 67 ████████████░░░░ 47% ║
║ patterns: 41 ██████████░░░░░░ 29% ║
║ prefs: 19 █████░░░░░░░░░░░ 13% ║
║ workflows: 15 ████░░░░░░░░░░░░ 11% ║
║ ║
║ Top Impact Learnings: ║
║ 1. JWT: use python-jose[crypto] (score 9) ║
║ 2. Docker: always set platform (score 8) ║
║ 3. Git: rebase before merge (score 7) ║
║ ║
║ Cross-Project Promotions: 37 ║
║ Archived (stale): 23 ║
║ Context Budget: 1,847 / 2,000 tokens ║
╚═══════════════════════════════════════════╝
learn promote — Force promotion check:
> learn promote
🔄 Running promotion check across all projects...
Scanning: web-api (67 learnings)
Scanning: ml-pipeline (41 learnings)
Scanning: mobile-app (34 learnings)
📊 Promotion Results:
✅ Promoted to global (score > 0.7):
- "Use --frozen-lockfile with npm ci in CI" (score: 0.82, seen in 4 projects)
- "Set timeout on all HTTP client calls" (score: 0.75, seen in 3 projects)
💡 Candidates (score 0.3-0.7):
- "Prefer pathlib over os.path in Python" (score: 0.55, seen in 2 projects)
🗄 Archived (score < 0.3 for 30+ days):
- "Pin Node.js to v16" (score: 0.12, stale 45 days)
Summary: 2 promoted, 1 candidate, 1 archived
learn export — Export all learnings as JSON:
> learn export
📦 Exporting all learnings...
Global learnings: 37 entries
Project learnings:
web-api: 67 entries
ml-pipeline: 41 entries
mobile-app: 34 entries
✅ Exported to self-learning-export-20260301.json
File size: 48 KB
Total entries: 179 (including 23 archived)
Format: Self-Learning Agent v1.0.0 portable JSON
learn import <file> — Import learnings:
> learn import team-learnings.json
📥 Importing from team-learnings.json...
Source: 215 entries (Self-Learning Agent v1.0.0 format)
Processing:
✅ New learnings added: 89
🔄 Merged with existing: 43 (frequency updated)
⏭ Duplicates skipped: 78
⚠ Conflicts detected: 5
Conflicts (require resolution):
1. "Use Jest" vs existing "Use Vitest" for testing framework
2. "Tabs for indentation" vs existing "Spaces for indentation"
... (3 more)
Run `learn promote` to resolve conflicts.
Import complete: 132 entries processed, 89 new.
learn prune — Clean up knowledge base:
> learn prune
🧹 Pruning knowledge base...
[1/4] Archiving stale learnings (unused >30 days)...
Archived: 12 entries
[2/4] Merging duplicates...
Merged: 8 pairs → 8 unified entries
Removed: 8 redundant entries
[3/4] Recalculating all scores...
Updated: 142 entries
Score changes: 23 entries rescored
[4/4] Compressing context...
Before: 2,847 tokens
After: 1,623 tokens
Saved: 1,224 tokens (43%)
✅ Prune complete
Entries: 162 → 142 (−20)
Disk saved: 12 KB
Context budget: 1,623 / 2,000 tokens
learn graph — Show knowledge graph:
> learn graph
🌐 Knowledge Graph
Category Tree:
├── errors (67)
│ ├── runtime (28)
│ ├── config (19)
│ ├── dependency (12)
│ └── permission (8)
├── patterns (41)
│ ├── workflow (18)
│ ├── coding (15)
│ └── deployment (8)
├── preferences (19)
│ ├── tooling (11)
│ └── style (8)
└── workflows (15)
├── CI/CD (9)
└── testing (6)
Cross-Project Connections:
web-api ←→ ml-pipeline: 12 shared learnings
web-api ←→ mobile-app: 8 shared learnings
ml-pipeline ←→ mobile-app: 3 shared learnings
Hub Nodes (most connected):
1. "Docker platform flag" → 4 projects, 7 uses
2. "npm ci --frozen-lockfile" → 4 projects, 6 uses
3. "Set HTTP timeout" → 3 projects, 5 uses
Total: 142 active | 23 archived | 37 global
Session Start:
.self-learning/config.jsonDuring Session:
Session End:
~/.openclaw/self-learning/
├── global/
│ ├── errors.jsonl # Cross-project errors
│ ├── patterns.jsonl # Universal patterns
│ ├── preferences.json # User preferences
│ └── index.json # Fast lookup index
├── projects/
│ └── {project-hash}/
│ ├── learnings.jsonl # Project learnings
│ ├── overrides.json # Project-specific config
│ └── archive.jsonl # Archived learnings
├── analytics/
│ └── stats.json # Aggregate statistics
└── config.json # Global config
| Feature | Self-Improving Agent | Self-Learning Agent |
|---|---|---|
| Cross-project memory | No (project-scoped) | Yes (global + project) |
| Automatic capture | Manual only | Auto + manual |
| Context management | Grows unbounded | 2000 token hard cap |
| Knowledge promotion | Manual status updates | Automatic scoring |
| Stale knowledge cleanup | None | 30-day auto-archive |
| Knowledge search | File-based | Indexed + scored |
| Import/Export | No | Yes |
| Deduplication | No | Automatic |
| Feature | Manual Notes | Mem0 | LangChain Memory | Self-Learning Agent |
|---|---|---|---|---|
| Cross-project memory | No (file-based) | Yes (cloud) | Yes (vector store) | Yes (local filesystem) |
| Automatic error capture | No | No | No | Yes (auto-capture on failure) |
| Knowledge promotion | Manual | Manual tags | Manual | Automatic scoring + promotion |
| Context budget control | Unbounded | Token-aware | Configurable | 2000 token hard cap + compression |
| Stale knowledge cleanup | Manual delete | Manual | Manual | 30-day auto-archive |
| Privacy | Local files | Cloud-hosted | Depends on store | Fully local, no network |
| Import/Export | Copy/paste | API export | Varies | Portable JSON format |
| Deduplication | Manual | Basic | No | Automatic with merge |
| Indexed search | No (text search) | Vector search | Vector search | Keyword index + scoring |
| Cost | Free (time cost) | $20+/mo | Free (self-hosted) | Free (zero token overhead for capture) |
| Conflict detection | No | No | No | Auto-detect + user resolution |
| Integration with evolution | No | No | No | Feeds into Capability Evolver Pro |
Q: How much context does it use? A: Maximum 2000 tokens per session, strictly enforced. Most sessions use 500-1500 tokens. The compression engine automatically merges, summarizes, and prioritizes learnings to stay within this budget.
Q: Does it slow down my agent? A: No. Captures happen asynchronously after tool calls. The only sync operation is loading learnings at session start (~100ms). Even with 500+ learnings in the knowledge base, the indexed lookup keeps query time under 50ms.
Q: Can I share learnings with my team?
A: Yes. Use learn export to create a portable JSON file, then learn import on another machine. The import process automatically deduplicates and merges with existing knowledge.
Q: How much does it cost in tokens?
A: The learning capture itself costs zero additional tokens — it piggybacks on existing tool call results. Loading context at session start costs 500-2000 tokens depending on the number of relevant learnings. The learn stats and learn graph commands each cost approximately 200-400 tokens for rendering.
Q: What happens when learnings conflict across projects? A: The promotion engine detects conflicts when a learning from Project A contradicts one from Project B. Conflicting learnings are flagged and presented to the user for resolution. Until resolved, both learnings remain at project level and neither is promoted to global.
Q: Can I use Self-Learning Agent with Capability Evolver Pro?
A: Yes, they complement each other well. Self-Learning Agent captures error patterns and knowledge, while Capability Evolver Pro acts on that knowledge to improve agent behavior. Evolver Pro can read Self-Learning Agent's learnings.jsonl as input for its repair strategy.
Q: Does it work offline? A: Fully offline. All knowledge storage is local filesystem-based (JSONL and JSON files). No network requests, no cloud sync, no external dependencies. The skill works entirely through standard file operations.
Q: How does auto-archive work and can I recover archived learnings?
A: Learnings with a promotion score below 0.3 for 30 consecutive days are automatically moved to .self-learning/archive/. Archived learnings are not loaded into context but remain on disk. Use learn recall <topic> to search across both active and archived learnings. You can manually re-activate an archived learning by moving it back to learnings.jsonl.
Q: Is my learning data private?
A: All data stays on your local machine in ~/.openclaw/self-learning/ (global) and <project>/.self-learning/ (project-level). No telemetry is collected. The learn export function creates a local file — sharing is entirely manual and user-initiated.
Q: How do I reset or start fresh?
A: Delete the .self-learning/ directory in the specific project, or ~/.openclaw/self-learning/ for global learnings. Alternatively, use learn prune to clean up stale learnings without a full reset. There is no learn reset command by design — the prune approach is safer and preserves high-value learnings.
| Error | Cause | Agent Action |
|---|---|---|
| Learning capture fails | Disk full or permissions issue on .self-learning/ directory | Log a warning to stderr. Do not interrupt the user's workflow — learning capture is non-blocking. Retry on next error event. Suggest checking disk space if failures persist. |
| JSONL parse error | Corrupted entry in learnings.jsonl or errors.jsonl | Skip the corrupted line, log its line number, and continue processing remaining entries. Suggest running learn prune to clean up corrupted records. |
| Context budget exceeded (>2000 tokens) | Too many high-scoring learnings loaded at session start | Apply compression strategies in order: merge similar → summarize verbose → prioritize by score → truncate lowest-score items. Never exceed the 2000-token hard cap. |
| Duplicate learning detected | Same error or pattern captured multiple times | Merge with existing learning: increment frequency, update created timestamp, recalculate promotion score. Do not create a duplicate entry. |
| Promotion conflict | Two project-level learnings contradict each other | Flag both learnings as conflicted status. Present both to user with context. Do not auto-promote either. Wait for user resolution via learn promote. |
| Index corruption | index.json is out of sync with actual learnings | Rebuild index from source JSONL files automatically. Log the rebuild event. This is a self-healing operation — no user action required. |
| Import merge failure | Imported file has incompatible schema or version | Report the specific incompatibility (missing fields, wrong version). Attempt partial import of compatible entries. Show count of skipped vs imported entries. |
| Global directory missing | ~/.openclaw/self-learning/global/ does not exist (first run) | Create the full directory structure with empty JSONL files, default config.json, and empty index.json. Log initialization event. Continue normally. |
| Archive directory full | Large number of archived learnings consuming disk space | Report disk usage of archive directory. Suggest running learn prune with --hard flag to permanently delete archived learnings older than 90 days. |
| Cross-project lookup timeout | Searching across many projects takes too long | Set a 5-second timeout on cross-project searches. Return partial results with a note indicating which projects were not searched. Suggest narrowing the search topic. |