aegntic/self-modify
crates/clawreform-skills/bundled/self-modify/SKILL.md
Enable ClawReform to safely modify its own codebase
$ install --global
skillsauthnpx skillsauth add aegntic/clawreform self-modifyInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Mar 30, 2026, 4:48 PM58.0s2 files scanned
SKILL.md
- name:
- self-modify
- description:
- Enable ClawReform to safely modify its own codebase
- version:
- 0.1.0
- author:
- ClawReform
- tags:
- [self-modification, code-generation, safety]
Self-Modification Skill
Enable ClawReform to safely modify its own codebase through natural language requests.
Capabilities
- Analyze natural language feature requests
- Create modification plans with risk assessment
- Apply code changes with automatic backups
- Validate changes via build/test/clippy
- Auto-rollback on validation failure
Usage
User: Add a /api/health/live endpoint for liveness probes
ClawReform will:
- Analyze the request and identify affected files
- Create a modification plan with risk level
- Backup affected files
- Apply the changes
- Run cargo build + test + clippy
- Rollback if validation fails
Safety Features
- Automatic Backups: All files backed up before modification
- Full Validation: Build, test, and clippy must pass
- Auto-Rollback: Failed validation triggers automatic restore
- Rate Limiting: Max 5 files per modification
- Approval System: High-risk changes require confirmation
API Endpoints
| Endpoint | Method | Purpose |
|----------|--------|---------|
| /api/self-modify/analyze | POST | Analyze request → create plan |
| /api/self-modify/apply | POST | Apply plan with validation |
| /api/self-modify/rollback | POST | Restore from backup |
| /api/self-modify/status | GET | Get current status |
| /api/self-modify/history | GET | Get modification history |
Examples
Add a New API Endpoint
"Add a /api/metrics endpoint that returns system metrics"
Create a New Skill
"Create a skill for summarizing PDF documents using the memory MCP server"
Fix a Bug
"Fix the memory leak in MCP connection handling"
Architecture
The self-modification system consists of:
- Analyzer: Parses natural language → identifies files, changes needed
- Backup Manager: Creates/restores file snapshots
- Modifier: Applies code changes safely
- Validator: Runs cargo build, test, clippy
- Rollback Handler: Restores from backup on failure
Constraints
- Maximum 5 files per modification
- Changes must pass all tests
- No modifications to .git directory
- No modifications to secrets/credentials
Related Skills
aegntic/writing-coach
testing
VerifiedTrustedCommunity
Writing improvement specialist for grammar, style, clarity, and structure
2SKILL.mdUpdated Mar 27, 2026
aegntic/web-search
development
VerifiedTrustedCommunity
Web search and research specialist for finding and synthesizing information
2SKILL.mdUpdated Mar 27, 2026
aegntic/wasm-expert
development
VerifiedTrustedCommunity
WebAssembly expert for WASI, component model, Rust/C compilation, and browser integration
2SKILL.mdUpdated Mar 27, 2026
aegntic/vector-db
data-ai
VerifiedTrustedCommunity
Vector database expert for embeddings, similarity search, RAG patterns, and indexing strategies
2SKILL.mdUpdated Mar 27, 2026