ship

Pass

Audited by Gen Agent Trust Hub on Aug 31, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The workflow involves running standard project verification commands like pnpm build, pnpm typecheck, pnpm lint, and pnpm test as pre-ship checks. These are expected behaviors for this skill's purpose in a development environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes multiple repository files that serve as ingestion points for untrusted data, especially in repositories with multiple contributors.
  • Ingestion points: The skill reads AGENTS.md, TASKS.md, and markdown files within the docs/ directory (e.g., docs/task/, docs/testing/, docs/learnings/).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in the workflow when reading these files.
  • Capability inventory: The skill can execute shell commands (pnpm), perform file system writes (updating TASKS.md), and trigger version control operations like branch and PR creation.
  • Sanitization: The instructions do not define methods to sanitize or validate the content of ingested files before they influence the agent's actions or the generated PR body.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 31, 2026, 07:05 AM
Security Audit — agent-trust-hub — ship