skill-authoring-workflow

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow relies on executing local shell and Python scripts (e.g., scripts/add-a-skill.sh, scripts/test-a-skill.sh, scripts/check-skill-metadata.py) to automate skill creation, validation, and packaging within the repository.
  • [INDIRECT_PROMPT_INJECTION]: The workflow involves processing external, potentially untrusted content such as workshop notes, transcripts, or frameworks to generate new skill assets, creating an entry point for indirect instructions.
  • Ingestion points: SKILL.md (Phase 2), template.md (Phase 2)
  • Boundary markers: Not explicitly specified for input source material.
  • Capability inventory: Local script execution (bash, python3) and filesystem write operations.
  • Sanitization: No specific input sanitization procedures are detailed before the content is processed by the authoring scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 04:11 PM
Security Audit — agent-trust-hub — skill-authoring-workflow