rapid-proto

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill's core workflow involves building and iterating on software based on user-provided data. \n
  • Ingestion points: The skill instructs the implementation of a FeedbackForm in SKILL.md and an analytics tracking system to capture external user input. \n
  • Boundary markers: The instructions and provided code examples lack boundary markers or warnings to the agent to treat captured feedback as untrusted content. \n
  • Capability inventory: The skill utilizes powerful tools including bash, git, and File ops as documented in the Copilot CLI Operations section. \n
  • Sanitization: There is no mention of sanitizing or validating the content of the feedback before it is used for the "Step 4: User Testing and Iteration Setup" where the agent is expected to make improvements based on that data.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 02:51 PM
Security Audit — agent-trust-hub — rapid-proto