skills/seabbs/skills/org-deps/Gen Agent Trust Hub

org-deps

Warn

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill explicitly instructs the agent to execute a shell script located at ~/.claude/scripts/org-deps.sh. Since the content of this script is not part of the skill definition, its specific actions and safety cannot be verified.
  • [COMMAND_EXECUTION]: In the 'Phase 4: Fix' section, the skill directs the agent to 'Run tests to verify compatibility'. Running test suites is a high-risk operation because tests can execute arbitrary code contained within the repositories being audited.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data from multiple repositories.
  • Ingestion points: The skill reads NEWS.md files and dependency metadata (R and Julia configuration files) from various repositories.
  • Boundary markers: No boundary markers or 'ignore embedded instructions' warnings are implemented to protect the agent from malicious instructions hidden within audited files.
  • Capability inventory: The agent is granted capabilities to execute local scripts (org-deps.sh), run arbitrary test code, and perform Git operations like creating branches and Pull Requests.
  • Sanitization: No sanitization or validation steps are defined for the content extracted from NEWS.md or other repository files before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 8, 2026, 07:16 AM