go-spec-reviewer
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted design documents (specs, RFCs, PRDs) which creates an attack surface where an attacker could embed malicious instructions to influence the agent's review or behavior.\n
- Ingestion points: The skill ingests data from external specification files provided at
[SPEC_FILE_PATH]and reads existing codebase files during its exploration phase.\n - Boundary markers: The instructions do not utilize explicit delimiters or "ignore embedded instructions" warnings when interpolating the spec content into the prompt.\n
- Capability inventory: The agent has the capability to read files from the filesystem, explore directory structures, and dispatch tasks to subagents.\n
- Sanitization: There is no evidence of sanitization, escaping, or schema validation for the untrusted content before it is processed.
Audit Metadata