skill-orchestrator

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The manifest defines several tools that execute local Python scripts (e.g., scripts/write_skills_index.py, scripts/suggest_skills.py) to manage project-specific skill indexing and documentation workflows. This behavior is consistent with the skill's stated purpose as an orchestrator.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface as it ingests untrusted data through parameters in its tools.
  • Ingestion points: Untrusted data enters via the user_prompt parameter in suggest_skills and the opinion parameter in append_pillar_opinion.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the manifest or the alias documentation.
  • Capability inventory: The skill possesses tools that run Python scripts with the ability to write to the local filesystem (e.g., write_skills_index, init_workflow_sync).
  • Sanitization: No sanitization or validation logic for user-provided input is specified in the provided files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 09:11 PM
Security Audit — agent-trust-hub — skill-orchestrator