sf-create-agentsmd
Pass
Audited by Gen Agent Trust Hub on Apr 18, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides templates to query official package registries such as PyPI, npm, and Packagist for metadata using curl. These operations target well-known technology services and are intended to verify dependency security and compatibility rather than execute remote code.- [PROMPT_INJECTION]: The skill ingests content from project files like README.md and Justfile to generate documentation, creating a surface for indirect prompt injection. Ingestion points: local project files identified during discovery (SKILL.md). Boundary markers: none detected. Capability inventory: documentation generation and project command classification (SKILL.md). Sanitization: no explicit filtering or escaping of ingested project content is performed.- [COMMAND_EXECUTION]: The skill facilitates security by requiring the categorization of project tasks into Safe, Dangerous, and Destructive tiers in the resulting AGENTS.md, providing clear execution guidelines and safety guardrails for agents.
Audit Metadata