skill-standardization

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill's instructions require the execution of several Python scripts (convert_skills.py, remove_duplicates.py, final_cleanup.py) to perform file modifications. These scripts are not provided in the skill package, making their internal operations and security posture unverifiable.
  • [PROMPT_INJECTION]: The skill is designed to ingest and process other SKILL.md files, establishing a surface for indirect prompt injection where malicious instructions within the processed data could influence the agent's behavior during standardization.
  • Ingestion points: Reads external SKILL.md files from the filesystem.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are identified in the provided procedure.
  • Capability inventory: The skill utilizes Bash, Read, Write, Edit, and Grep tools, allowing for significant filesystem and command-line interaction.
  • Sanitization: There is no evidence of sanitization or content validation of the markdown files being processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 07:53 AM