reframe
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-generated markdown documents within the
concepts/directory to manage design state and generate retrospectives. This creates a surface for indirect prompt injection where malicious instructions in those files could influence agent behavior.\n - Ingestion points: The skill reads
concepts/*.mdfiles in both the main workflow and the/reframe closecommand (Step 2).\n - Boundary markers: Content from concept documents is read into the context without explicit boundary markers or instructions to ignore embedded instructions.\n
- Capability inventory: The skill directs the agent to read, write, and move files, and to perform git commits.\n
- Sanitization: No sanitization or validation is performed on the ingested markdown content before it is processed by the agent.
Audit Metadata