Skill of Skills
📄 skillStaleExperimentalmedium

run-llama/agentfs-claude

by run-llama

Run Claude Code/Codex within AgentFS, orchestrated by LlamaIndex Workflows

Supported Platforms

🧠 Codex

Stars

324

Skill Type

📖 Library & API Reference

Quality Score

62/200

License

MIT

Forks

31

Last Updated

Jul 24, 2026

Discovered

May 11, 2026

Validation

Passed

github.com/run-llama/agentfs-claude

Quality Breakdown

62/ 200

Content Signals

○Gotchas/Edge Cases+40
○Progressive Disclosure+30
○Trigger Description+20
○Verification/Safety+20
✓Code Examples+15
✓Composability+15

Repo Health

○Recent Activity+15
○Scripts/Automation+10
○Real Usage (Issues)+10
✓Single Responsibility+10
✓Config/Persistence+10
✓Install Instructions+5

Multi-platform bonus: +5 pts if tool supports 2+ platforms. Score derived from 12 structural signals — not stars or popularity.

Trust & Verification

medium

Requires extended permissions (shell access, subagents). Review before use.

Stale

No commits in 90-365 days and no tagged releases.

Unverified skill. Always review source code before installing any skill from an unknown author.

Risk Assessment

  • MCP (Model Context Protocol) server implementation that exposes filesystem operations (read_file, write_file, edit_file, list_files) - potential for unauthorized file access
  • LlamaIndex Workflows orchestration of Claude Code/Codex agents - multi-agent spawning capability
  • MCP server runs on localhost:3000 and requires global Codex configuration modification (~/.codex/config.toml)
  • Integration with external cloud service (LlamaCloud) for document processing
  • Codex SDK installation with large dependencies (140+ MB) and separate MCP server startup process
  • Agent autonomy to read/write/edit files within virtualized AgentFS (though virtualized, reduces friction for unauthorized operations)

Related guides

Best skills for OpenAI Codex