convex-sentinel

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing the @convex-dev/sentinel package from the official vendor namespace to facilitate production error monitoring within the Convex ecosystem.- [PROMPT_INJECTION]: The skill describes a workflow that ingests production error logs, creating an indirect prompt injection surface.
  • Ingestion points: Server function failures and client JS/React crashes recorded in the sentinel table.
  • Boundary markers: None explicitly defined for LLM interpolation, though redaction is enforced.
  • Capability inventory: Triggers ai-runner to open fix Pull Requests.
  • Sanitization: Default-deny redaction for secret keys and patterns at write time.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 03:15 PM
Security Audit — agent-trust-hub — convex-sentinel