api-attack-surface-mapper
Pass
Audited by Gen Agent Trust Hub on Oct 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because its core workflow involves processing potentially untrusted external data sources.
- Ingestion points: Data enters the agent context through the
api_spec_source(OpenAPI URLs, Postman collections) and "captured traffic" during the input normalization and trust-boundary mapping phases (SKILL.md, Execution Workflow). - Boundary markers: The instructions do not define explicit delimiters or boundary markers to separate external data from system instructions, nor do they include warnings to disregard instructions embedded within the processed data.
- Capability inventory: The skill utilizes network capabilities in Phase 5 ("Low-Noise Validation") to confirm endpoint liveness and record response fingerprints (SKILL.md, Phase 5).
- Sanitization: No sanitization, escaping, or schema validation steps are specified to handle the content of the external specifications or traffic before it is processed by the agent.
- [NO_CODE]: The skill consists exclusively of markdown instructions (SKILL.md) and does not include any accompanying scripts, executables, or configuration files that would require code-level analysis.
Audit Metadata