pre-ship-gate

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides example shell commands for comparing local git revisions with live service data using git rev-parse and bash logic.\n- [EXTERNAL_DOWNLOADS]: Includes instructions and examples for fetching deployment status and revision identifiers from remote health endpoints via curl.\n- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by consuming data from external status endpoints.\n
  • Ingestion points: Remote status and health endpoints (e.g., https://your-service.example.com/health) as described in the verification phase of SKILL.md.\n
  • Boundary markers: No specific delimiters or warnings are used to isolate the data fetched from external endpoints.\n
  • Capability inventory: The skill utilizes curl for network requests, jq for data processing, and git for local repository state checking.\n
  • Sanitization: External content is filtered through jq to extract specific fields like '.revision', which acts as a basic structural filter but does not sanitize the string content.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 08:47 PM
Security Audit — agent-trust-hub — pre-ship-gate