miro-ci-integration

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill emphasizes security best practices, such as isolating credentials from fork/PR workflows and redacting sensitive information from test evidence. It operates within expected tool constraints for repository management.- [EXTERNAL_DOWNLOADS]: The skill references official Miro documentation, developer guides, and status pages. These references are to well-known service endpoints and are used solely for informational purposes.- [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill reads repository content, including adapters, tests, and evidence files, using tools like Read, Glob, and Grep.
  • Boundary markers: The skill instructions do not specify technical delimiters but enforce manual approval boundaries and operator-led live execution.
  • Capability inventory: The skill permits the use of Write and Edit tools for repository-side implementation.
  • Sanitization: Detailed instructions are provided to redact identifiers and avoid printing access tokens, refresh tokens, or secrets.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 11:34 PM
Security Audit — agent-trust-hub — miro-ci-integration