research
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process implementation details from external sources such as documentation sites and maintainer repositories. This functionality presents a surface for indirect prompt injection where malicious instructions embedded in a third-party webpage could attempt to influence the agent's behavior during synthesis.
- Ingestion points: Web documentation, API references, and GitHub repositories fetched via web tools (SKILL.md, Step 3).
- Boundary markers: None; the instructions do not specify the use of delimiters or 'ignore' instructions when processing the external content.
- Capability inventory: Reads local manifests/configuration files (SKILL.md, Step 2) and writes synthesized Markdown files to the local repository (SKILL.md, Step 4).
- Sanitization: None; the process lacks explicit sanitization or filtering of the retrieved web content before it is incorporated into the agent's context.
Audit Metadata