research
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a research process where an agent is tasked with fetching information from untrusted external sources (such as documentation, source code, and APIs) and writing the findings to a Markdown file within the local repository. This creates a vulnerability to indirect prompt injection, where an attacker could place malicious instructions within those external sources to influence the agent's behavior or poisoning the repo content.
- Ingestion points: External primary sources (official docs, source code, specs, APIs) as specified in SKILL.md.
- Boundary markers: The instructions do not define any specific delimiters or warnings to ignore instructions embedded within the fetched data.
- Capability inventory: The skill utilizes the agent's capability to write files to the repository filesystem.
- Sanitization: The skill does not mention or require any validation, filtering, or escaping of the content gathered from external sources.
- [NO_CODE]: The skill is comprised entirely of natural language instructions and configuration metadata. It does not contain any scripts, binaries, or automated installation steps.
Audit Metadata