api-state
Pass
Audited by Gen Agent Trust Hub on Sep 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources and processes it to generate technical reports, creating an attack surface for indirect prompt injection.\n- Ingestion points: External content is retrieved in SKILL.md (Step 2 and Step 3) using tools such as
mcp__Ref__ref_read_urlandWebFetch.\n- Boundary markers: The skill instructions lack delimiters or specific directives to ignore instructions that might be embedded within the retrieved documentation or changelogs.\n- Capability inventory: The skill's capabilities are limited to information retrieval and reporting; it does not have access to sensitive file systems, shell execution, or stateful modifications.\n- Sanitization: No sanitization or validation mechanisms are defined to filter the content of the external sources before analysis.
Audit Metadata