pythia-skill-author
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill utilizes a suite of CLI tools (e.g.,
pythia similar,pythia src,pythia skill new,pythia skill lint) to perform codebase analysis, signature extraction, and file scaffolding. These commands are integral to the skill's primary purpose of capturing and verifying development workflows within the local environment. - [INDIRECT_PROMPT_INJECTION]: The skill has an ingestion surface where it reads verbatim snippets from the local codebase (Step 2) to use as examples in the new skill.
- Ingestion points: Source code files accessed via
pythia similarandpythia srcinSKILL.md. - Boundary markers: None explicitly defined for the mined code snippets.
- Capability inventory: File system reading (
pythia src) and writing/scaffolding (pythia skill new). - Sanitization: No automated sanitization is mentioned, though the workflow includes a manual drafting phase by the agent and verification by the developer.
Audit Metadata