research
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from external "primary sources" (official docs, source code, specs, first-party APIs) which are untrusted external inputs. This content is then processed and written to the local repository. An attacker who controls one of these sources could embed instructions to manipulate the agent's output or perform unauthorized actions.
- Ingestion points: External primary sources, official documentation, and first-party APIs (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters used to prevent the agent from following commands found within the external sources.
- Capability inventory: The skill utilizes file-writing capabilities to save findings to the repository (SKILL.md).
- Sanitization: The skill does not specify any sanitization, filtering, or validation of the external content before it is incorporated into the generated Markdown files.
Audit Metadata