miro
Pass
Audited by Gen Agent Trust Hub on Apr 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses
curlandjqto execute API requests and process JSON responses. It also referenceszero doctor, a vendor-specific diagnostic tool for troubleshooting connection issues. - [DATA_EXFILTRATION]: Performs network operations targeting
api.miro.com, which is the official API for Miro, a well-known visual collaboration service. - [PROMPT_INJECTION]: The skill processes data from Miro boards (such as sticky note content and board details), representing an attack surface for indirect prompt injection.
- Ingestion points: Board and item data retrieved from Miro API endpoints in
SKILL.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish between its instructions and retrieved board content.
- Capability inventory: Shell command execution via
curlandjq, and the ability to write temporary files to/tmp/miro_request.json. - Sanitization: Uses
jqfor structural extraction of JSON fields, but does not sanitize the actual text content of items retrieved from the API.
Audit Metadata