miro-code-review
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill analyzes external data from pull/merge requests (diffs and descriptions), creating a potential surface for indirect prompt injection where malicious instructions embedded in the reviewed code could attempt to manipulate the agent's output.
- Ingestion points: Fetches PR/MR title, body, and unified diff from GitHub/GitLab via platform CLIs or REST API as described in Step 2 of SKILL.md.
- Boundary markers: The instructions do not define clear boundaries or delimiters to help the agent distinguish between its operational instructions and the potentially instructional content of the files or descriptions it is reviewing.
- Capability inventory: The agent can create and modify Miro board items (tables, documents, and diagrams) and update PR/MR descriptions or post comments on forges as detailed in Step 5 and 6 of SKILL.md.
- Sanitization: No explicit sanitization or filtering of the retrieved diff content is described before it is analyzed or used to generate summaries and diagrams.
Audit Metadata