miro-spec-guide
Pass
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: Indirect prompt injection vulnerability surface.
- Ingestion points: Content from Miro boards is fetched via URLs and stored locally in the
.miro/specs/directory (SKILL.md). - Boundary markers: The skill does not specify the use of delimiters or 'ignore embedded instruction' warnings when the agent reads the extracted files.
- Capability inventory: The agent is encouraged to use the extracted content for implementation planning and design, which requires interpreting data from potentially untrusted Miro boards (SKILL.md).
- Sanitization: There is no mention of sanitizing or filtering the content extracted from the Miro board before it is processed by the agent.
Audit Metadata