shunk031-research-before-implementation
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill directs the agent to fetch and read content from external documentation sites and GitHub repositories, which are attacker-controlled sources.
- Ingestion points:
SKILL.mdinstructs the agent to use web-research capabilities to read source content from third-party documentation and GitHub implementation code. - Boundary markers: The skill lacks instructions for the agent to treat this external content as untrusted or to use delimiters to prevent embedded instructions from influencing its behavior.
- Capability inventory: The skill is intended to be used in a context where the agent has the capability to implement and verify changes, implying file system and execution permissions.
- Sanitization: No validation or sanitization of the retrieved external content is specified before the agent processes it.
Audit Metadata