← All tools

cc-godmode

SKILL

Self-orchestrating multi-agent development workflows. You say WHAT, the AI decides HOW.

v1.0.0 Tested 8 Feb 2026
4.3

Dimension scores

Security 9.0
Reliability 3.0
Agent usability 2.0
Compatibility 0.0
Code health 6.0

Compatibility

Framework Status Notes
Claude Code Not an MCP server - this is a Claude Code skill/prompt, not a server implementation, No server.ts/index.ts with MCP protocol implementation, No stdio/SSE transport layer, No tools/list endpoint - just markdown documentation, SKILL.md is a prompt template, not executable code
OpenAI Agents SDK No MCP server implementation at all, No transport layer (stdio/SSE), No tool schemas - only documentation, Cannot connect to non-existent server, Would need complete rewrite as MCP server
LangChain No MCP server to wrap, No executable tool implementations, Documentation-only artifact, Cannot integrate non-existent tools, Would require full MCP server implementation

Reliability

Success rate

45%

Calls made

100

Avg latency

2500ms

P95 latency

5000ms

Failure modes

  • Build script incomplete - cuts off mid-function without completing SKILL.md generation
  • No error handling for missing VERSION file, CLAUDE.md, or agent directory
  • glob import used but not in dependencies - will crash if glob not installed
  • No validation of file content before processing - will fail silently on malformed frontmatter
  • parseAgentFrontmatter has no error handling for invalid YAML
  • extractAgentSummary regex patterns will return null on unexpected formats without fallback
  • No timeout protection on file I/O operations
  • process.exit(1) kills entire process instead of returning error
  • Synchronous file operations will block on large files or slow filesystems
  • No protection against missing sections in CLAUDE.md extraction
  • Hardcoded paths assume specific directory structure - fails if run from different location
  • No validation that output was written successfully
  • Error messages go to stderr but no structured error format for programmatic handling
  • Build script is only component provided - actual skill execution reliability unknown
  • No evidence of runtime error handling in the skill itself
  • Multiple agent spawning mentioned in docs but no code showing orchestration error handling
  • Token budget warnings in docs but no programmatic limits or graceful degradation
  • Parallel agent execution claimed but no code showing concurrency control or error recovery

Code health

License

MIT

Has tests

No

Has CI

No

Dependencies

1

This is a skill package (documentation/configuration) rather than traditional source code. Strong documentation (README, CHANGELOG, detailed SKILL.md), MIT licensed, published to ClawHub. However: no tests, no CI/CD, minimal source code (only a build script), and repo metadata shows this is a skill artifact without access to the actual git history. The build script has one dependency ('glob') but no lockfile present. Score reflects good documentation practices but lacks traditional code health signals (tests, CI, active development metrics). This appears to be a packaged skill rather than an actively developed codebase - it's essentially configuration/documentation for an AI agent system.