tiangolo-library-skills
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the download and execution of the
library-skillspackage from the official PyPI and npm registries using theuvxandnpxrunners. This is the intended behavior for tool installation.\n- [INDIRECT_PROMPT_INJECTION]: The skill creates a surface for indirect prompt injection by instructing the agent to discover and adhere to instructions provided by third-party libraries installed in the project environment.\n - Ingestion points: Untrusted instructions enter the agent context via
SKILL.mdfiles located within the package directories of installed dependencies (e.g.,fastapi/skills/SKILL.md) and metadata from theagentskills.ioregistry.\n - Boundary markers: Absent. The skill instructions encourage the agent to adopt the patterns found in these external files without providing delimiters or warnings to ignore malicious instructions embedded within them.\n
- Capability inventory: The skill environment allows for environment scanning, directory manipulation (creating symlinks in
.agents/), and shell command execution via package managers.\n - Sanitization: No sanitization, validation, or content filtering is described for the instructions retrieved from external libraries before they are applied by the agent.
Audit Metadata