verify-api-shape
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill describes a procedure for capturing and analyzing live responses from external third-party APIs. This creates an attack surface where an attacker-controlled API response could contain malicious instructions designed to influence the agent's subsequent behavior or analysis.
- Ingestion points: The skill instructs the agent to "Capture a real response first" by running GET requests against external endpoints (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat the captured API data as untrusted text to be ignored in terms of instructions.
- Capability inventory: The skill provides Javascript logic for deep-searching objects and tallying field presence, but does not include dangerous capabilities like file system writes or unauthorized network exfiltration within the provided snippets (SKILL.md).
- Sanitization: The skill mentions that captured responses should be "sanitised" before being used to rebuild test fixtures (SKILL.md, Step 8), which is a recommended security practice.
Audit Metadata