miro-sdk-patterns

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by fetching documentation from external sources and reading local repository files, which could potentially contain adversarial content designed to influence agent behavior.
  • Ingestion points: The skill retrieves external documentation using WebFetch and inspects repository content using Read, Glob, and Grep (SKILL.md).
  • Boundary markers: Includes instructions to confirm targets and approval boundaries before performing modifications, and requires pausing when the owner or target is uncertain.
  • Capability inventory: The skill is authorized to use Read, Glob, Grep, WebFetch, Write, and Edit tools.
  • Sanitization: Employs proactive sanitization measures, including mandatory redaction of board content and credentials, and the use of structural diffs for evidence logging.
  • [EXTERNAL_DOWNLOADS]: Fetches documentation and implementation guidelines from official Miro developer resources.
  • Resource: https://developers.miro.com/docs/miro-nodejs-api-client (SKILL.md, references/official-docs.md)
  • Resource: https://developers.miro.com/reference/overview (SKILL.md, references/official-docs.md)
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 11:34 PM
Security Audit — agent-trust-hub — miro-sdk-patterns