Skills
Navigate
skills/block/agent-skills/beads/Gen Agent Trust Hub

beads

Pass

Audited by Gen Agent Trust Hub on Sep 20, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest task information from external sources (issue titles, descriptions, and notes), which provides a surface for indirect prompt injection. Ingestion points: bd ready, bd list, bd show in SKILL.md. Boundary markers: Absent. Capability inventory: git pull, git push, npm install, pip install, bd sync in SKILL.md. Sanitization: Absent.
  • [EXTERNAL_DOWNLOADS]: The skill suggests installation of the beads tool and MCP server from standard repositories including npm, PyPI, and GitHub.
  • [COMMAND_EXECUTION]: The skill defines workflows involving the execution of CLI tools for issue tracking and version control.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 20, 2026, 01:56 AM