🎉 init: 小龙的工作空间
This commit is contained in:
大海
@@ -0,0 +1,86 @@
|
||||
---
|
||||
name: code-explorer
|
||||
description: Search and explore codebases with CodeGraph. Symbol lookup, caller/callee analysis, impact analysis, BFS graph traversal.
|
||||
metadata: { "openclaw": { "emoji": "🔍" } }
|
||||
---
|
||||
|
||||
# Code Explorer — CodeGraph Knowledge Graph
|
||||
|
||||
Query codebases through a pre-indexed SQLite knowledge graph with FTS5 full-text search and BFS graph traversal.
|
||||
|
||||
## When to Use
|
||||
|
||||
- **"Where is X defined?"** → `search`
|
||||
- **"Who calls X?"** → `callers`
|
||||
- **"What does X call?"** → `callees`
|
||||
- **"What's the impact of changing X?"** → `impact`
|
||||
- **"How does the codebase handle auth?"** → `search` + `callers`
|
||||
- **Before grep/glob** — CodeGraph is faster and more structured
|
||||
|
||||
## Setup
|
||||
|
||||
```bash
|
||||
# Index a codebase (only needed once, or after major changes)
|
||||
node src/codegraph/index.js index <project_dir>
|
||||
|
||||
# Force re-index
|
||||
node src/codegraph/index.js index . --force
|
||||
|
||||
# Check index stats
|
||||
node src/codegraph/index.js stats
|
||||
```
|
||||
|
||||
## Commands
|
||||
|
||||
### Search symbols
|
||||
|
||||
```bash
|
||||
node src/codegraph/index.js search "function_name"
|
||||
node src/codegraph/index.js search "auth" --limit 10
|
||||
```
|
||||
|
||||
### Caller/callee analysis
|
||||
|
||||
```bash
|
||||
node src/codegraph/index.js callers "login"
|
||||
node src/codegraph/index.js callees "main"
|
||||
```
|
||||
|
||||
### Impact analysis (BFS traversal)
|
||||
|
||||
```bash
|
||||
node src/codegraph/index.js impact "AuthService"
|
||||
```
|
||||
|
||||
### Deep exploration (search + BFS context)
|
||||
|
||||
```bash
|
||||
node src/codegraph/index.js explore "how to send HTTP requests"
|
||||
```
|
||||
|
||||
## Languages Supported
|
||||
|
||||
JavaScript/TypeScript, Python, Rust, Go, Java, Ruby, Shell, C/C++
|
||||
|
||||
Extraction via regex grammars (8 languages, 8+ node types: function/class/method/struct/enum/import/export)
|
||||
|
||||
## Database
|
||||
|
||||
SQLite at `.codegraph/codegraph.db` with:
|
||||
- `nodes` — symbols (name, kind, signature, location)
|
||||
- `edges` — relationships (calls, imports, extends, implements)
|
||||
- `files` — file metadata with hash-based incremental sync
|
||||
- `nodes_fts` — FTS5 full-text search index
|
||||
|
||||
## Integration Pattern
|
||||
|
||||
```
|
||||
Query codebase → codegraph search → symbol list → codegraph callers → call graph
|
||||
↓
|
||||
codegraph impact → affected files
|
||||
```
|
||||
|
||||
## File Path
|
||||
|
||||
- `src/codegraph/index.js` — main entry (index, search, callers, callees, impact, explore)
|
||||
- `src/codegraph/schema.sql` — SQLite schema
|
||||
@@ -0,0 +1,84 @@
|
||||
---
|
||||
name: compression
|
||||
description: Compress tool outputs, logs, and JSON arrays before they enter LLM context. Saves 60-95% tokens.
|
||||
metadata: { "openclaw": { "emoji": "📦" } }
|
||||
---
|
||||
|
||||
# Compression — Context-Aware Output Compression
|
||||
|
||||
Compress large tool outputs before they consume LLM context. Content-type-aware: JSON arrays, build logs, search results, git diffs each get optimal compression.
|
||||
|
||||
## When to Use
|
||||
|
||||
- Tool output > 500 chars → compress before reading
|
||||
- JSON arrays > 10 items → SmartCrusher
|
||||
- Build/CI logs > 50 lines → template dedup
|
||||
- Search/grep results > 30 lines → file dedup
|
||||
- Git diffs with lock files → noise filter
|
||||
- Need to retrieve original compressed data → CCR store
|
||||
|
||||
## Tools
|
||||
|
||||
### `ctx_compress` — one-stop compression
|
||||
|
||||
```bash
|
||||
# Auto-detect type, compress stdin
|
||||
cat huge_output.json | node src/compression/ctx-compress.js
|
||||
|
||||
# Specify type, bias
|
||||
cat build.log | node src/compression/ctx-compress.js --type log --json
|
||||
|
||||
# Force JSON mode with query context for relevance
|
||||
cat results.json | node src/compression/ctx-compress.js --type json --query "error timeout" --json
|
||||
|
||||
# Retrieve original from CCR store
|
||||
node src/compression/ctx-compress.js --retrieve <ccr_hash>
|
||||
```
|
||||
|
||||
### `smart-crusher` — direct JSON compression
|
||||
|
||||
```bash
|
||||
node src/compression/smart-crusher.js --bias 0.7 --query "search terms" < input.json
|
||||
```
|
||||
|
||||
### Content type detection
|
||||
|
||||
```bash
|
||||
cat unknown_output.txt | node src/compression/detector.js
|
||||
```
|
||||
|
||||
## Compression Strategies
|
||||
|
||||
| Content Type | Strategy | Savings |
|
||||
|-------------|----------|:-------:|
|
||||
| JSON array | lossless:csv or smart sample | 50-90% |
|
||||
| Build log | template dedup | 90-99% |
|
||||
| Search results | file dedup | 80-95% |
|
||||
| Git diff | noise filter (lock files etc) | 30-70% |
|
||||
|
||||
## Key Feature: CCR (Compress-Cache-Retrieve)
|
||||
|
||||
Compressed output includes `<<ccr:HASH N_rows_offloaded>>` markers. If LLM needs original data, retrieve it:
|
||||
|
||||
```bash
|
||||
node src/compression/ctx-compress.js --retrieve <hash>
|
||||
```
|
||||
|
||||
## Integration Pattern
|
||||
|
||||
```
|
||||
large_tool_output → ctx_compress → compressed → LLM context
|
||||
↓
|
||||
<<ccr:abc123>>
|
||||
↓
|
||||
LLM needs more? → ctx_compress --retrieve abc123
|
||||
```
|
||||
|
||||
## File Paths
|
||||
|
||||
- `src/compression/ctx-compress.js` — main entry
|
||||
- `src/compression/smart-crusher.js` — JSON compression engine
|
||||
- `src/compression/detector.js` — content type detection
|
||||
- `src/compression/ccr-store.js` — reversible storage
|
||||
- `src/compression/ccr-backends.js` — SQLite/Redis persistence
|
||||
- `src/compression/byte-surgery.js` — cache-safe byte replacement
|
||||
@@ -0,0 +1,56 @@
|
||||
---
|
||||
name: security-check
|
||||
description: Pre-execution risk assessment for tool calls. Checks file sensitivity, blast radius, irreversibility before running dangerous commands.
|
||||
metadata: { "openclaw": { "emoji": "🛡️" } }
|
||||
---
|
||||
|
||||
# Security Check — Tool Risk Assessment
|
||||
|
||||
Run risk assessment before executing potentially dangerous operations. Four-factor weighted scoring adapted from ECC.
|
||||
|
||||
## When to Use
|
||||
|
||||
- **Before Bash/exec** — always check risk
|
||||
- **Before Write/Edit** — check file sensitivity
|
||||
- **Before git push --force / rm -rf** — mandatory check
|
||||
- **Before database DROP/TRUNCATE** — mandatory check
|
||||
- **Before chmod/chown** — check blast radius
|
||||
|
||||
## Tool
|
||||
|
||||
```bash
|
||||
# JSON input mode (preferred)
|
||||
node src/security/risk-scorer.js '{"tool":"Bash","command":"rm -rf /tmp/build"}'
|
||||
|
||||
# Simple mode
|
||||
node src/security/risk-scorer.js Write .env.production
|
||||
```
|
||||
|
||||
## Risk Levels
|
||||
|
||||
| Score | Action | Meaning |
|
||||
|:-----:|:------:|---------|
|
||||
| < 0.35 | **ALLOW** | Safe operation |
|
||||
| 0.35-0.60 | **REVIEW** | Review input before executing |
|
||||
| 0.60-0.85 | **CONFIRM** | Ask user for confirmation |
|
||||
| ≥ 0.85 | **BLOCK** | Refuse to execute |
|
||||
|
||||
## Four Factors
|
||||
|
||||
1. **Base tool risk** — Bash=0.20, Write=0.15, Read=0.02
|
||||
2. **File sensitivity** — .env +0.25, /etc/ +0.20, SSH keys +0.25
|
||||
3. **Blast radius** — rm -rf +0.35, curl-to-shell +0.25, SQL DROP +0.30
|
||||
4. **Irreversibility** — git push --force +0.45, hard reset +0.35
|
||||
|
||||
## Integration Pattern
|
||||
|
||||
```
|
||||
Before exec/write → risk-scorer → ALLOW → execute
|
||||
→ REVIEW → show risk factors
|
||||
→ CONFIRM → ask user
|
||||
→ BLOCK → refuse, explain why
|
||||
```
|
||||
|
||||
## File Path
|
||||
|
||||
- `src/security/risk-scorer.js` — scoring engine
|
||||
@@ -0,0 +1,84 @@
|
||||
---
|
||||
name: self-evolution
|
||||
description: Post-task reflection, pattern extraction, skill generation, memory update. Continuous self-improvement loop.
|
||||
metadata: { "openclaw": { "emoji": "🧬" } }
|
||||
---
|
||||
|
||||
# Self-Evolution — Continuous Improvement Loop
|
||||
|
||||
After every completed task, run the evolution cycle. Each task makes the system stronger.
|
||||
|
||||
## Evolution Cycle (9 Steps)
|
||||
|
||||
### 1. Record Successes
|
||||
What worked? Why? Can we replicate it?
|
||||
|
||||
→ `/reflections/YYYY-MM-DD-<task>.md`
|
||||
|
||||
### 2. Record Failures
|
||||
What broke? Root cause? Fix applied?
|
||||
|
||||
→ same reflection file
|
||||
|
||||
### 3. Generate New Skill
|
||||
Does this task create a reusable capability? If yes:
|
||||
|
||||
→ `/skills/<skill-name>.md` (YAML frontmatter + usage guide)
|
||||
|
||||
### 4. Update Memory
|
||||
Write today's key events to daily log.
|
||||
|
||||
→ `memory/daily/YYYY-MM-DD.md`
|
||||
|
||||
### 5. Update Best Practices
|
||||
Extract reusable patterns that worked.
|
||||
|
||||
→ `/patterns/design-patterns.md` (problem → solution → anti-pattern)
|
||||
|
||||
### 6. Update Prompt Templates
|
||||
Did we create a useful prompt format? Template it.
|
||||
|
||||
→ `/patterns/prompt-templates.md`
|
||||
|
||||
### 7. Update Tool Strategy
|
||||
Did we learn about tool selection/sequencing?
|
||||
|
||||
→ `/playbooks/` or pattern file
|
||||
|
||||
### 8. Update CodeGraph
|
||||
Re-index the workspace to capture new code.
|
||||
|
||||
```bash
|
||||
node src/codegraph/index.js index . --force
|
||||
```
|
||||
|
||||
### 9. Update Compression Rules
|
||||
Did we discover new content type → compression strategy mappings?
|
||||
|
||||
→ update `src/compression/detector.js` signal weights
|
||||
|
||||
## Directory Structure
|
||||
|
||||
```
|
||||
workspace/
|
||||
├── reflections/ ← 每次任务的反思 (YYYY-MM-DD-<task>.md)
|
||||
├── patterns/ ← 可复用设计模式
|
||||
├── playbooks/ ← 常用任务的操作手册
|
||||
├── skills/ ← 能力技能文件
|
||||
└── memory/ ← 长期记忆系统
|
||||
```
|
||||
|
||||
## Integration with Existing Systems
|
||||
|
||||
- **Dream Cycle** (cron 4:00) → 自动整合 reflections/ 到 memory/
|
||||
- **Memory Search** → 可检索过往反思
|
||||
- **CodeGraph** → 索引所有演化产物
|
||||
|
||||
## Quality Gate
|
||||
|
||||
After each evolution cycle, verify:
|
||||
- [ ] reflections/ 有本次任务的反思文件
|
||||
- [ ] 如果产生了新模式 → patterns/ 已更新
|
||||
- [ ] 如果产生了新能力 → skills/ 已创建
|
||||
- [ ] memory/daily/ 已写入今日总结
|
||||
- [ ] 所有新代码可独立运行
|
||||
Reference in New Issue
Block a user