skills/nearform/unwind/uw-verify/Gen Agent Trust Hub

uw-verify

Pass

Audited by Gen Agent Trust Hub on Oct 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Executes a bash script block that locates its own installation directory and invokes a Node.js process to perform a coverage analysis.
  • [DYNAMIC_EXECUTION]: Runs a local JavaScript file (verify-coverage.mjs) using the node runtime provided in the environment.
  • [INDIRECT_PROMPT_INJECTION]: Ingests documentation files and source code entry points to detect gaps. Ingestion points: reads project markdown files and source code listed in architecture.md. Boundary markers: none. Capability inventory: limited to file read/write and restricted bash commands. Sanitization: no explicit sanitization for manual LLM comparison.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 30, 2026, 01:33 AM