knock-migrate-to-knock

Pass

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill serves a legitimate discovery purpose, scanning a repository for messaging provider integration patterns to produce an inventory and migration plan in a local knock-plan.md file.
  • [SAFE]: It contains hard constraints against mutating external Knock resources (workflows, users, etc.) without explicit user consent.
  • [SAFE]: All external resources are links to official vendor documentation (docs.knock.app).
  • [PROMPT_INJECTION]: The skill inherently possesses an indirect prompt injection surface as it is designed to process untrusted repository data (e.g., templates and configuration manifests). Ingestion points: Repository search operations defined in rules/inventory-existing-messaging.md. Boundary markers: Absent. Capability inventory: File system search and file write access (knock-plan.md). Sanitization: Absent. This surface is considered an expected operational risk for a code analysis tool and does not indicate malicious intent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 24, 2026, 04:01 PM
Security Audit — agent-trust-hub — knock-migrate-to-knock