mcp-workflow-design
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructions define a process that involves ingesting untrusted data from the web.
- Ingestion points: SKILL.md Phase 3 explicitly requires the agent to perform web searches of vendor documentation, community forums, Stack Overflow, and integration partner blogs to identify user flows.
- Boundary markers: There are no instructions or delimiters provided to help the agent distinguish between legitimate data and potential malicious instructions embedded in these external sources.
- Capability inventory: The workflow relies on the agent's ability to search and read external web content, which then directly informs the design and implementation of new tools.
- Sanitization: The skill does not prescribe any validation or sanitization steps for content retrieved from external community sources.
Audit Metadata