reuse
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists entirely of markdown instructions and configuration metadata. There are no executable scripts or binaries included, which significantly reduces the attack surface.- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and analyze project source code, which is considered untrusted external data. This creates a surface where instructions could be embedded in code to influence the agent's behavior.
- Ingestion points: The agent is instructed to search the shared component package, icon sets, tokens, and existing call sites within the codebase (SKILL.md).
- Boundary markers: The instructions do not provide specific delimiters or warnings to ignore instructions found within the analyzed files.
- Capability inventory: The agent is empowered to Approve or Block reviews based on its findings and is instructed to patch primitive components (SKILL.md).
- Sanitization: No sanitization or validation of the ingested code content is defined in the instructions.
Audit Metadata