ship-autonomous

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources as part of its core functionality.
  • Ingestion points: Linear ticket descriptions and comments (Step 2), and GitHub PR review comments and reactions (Step 6).
  • Boundary markers: No specific delimiters or safety instructions are defined to separate user intent from external data content.
  • Capability inventory: The skill allows filesystem modifications, shell command execution (git, gh, gt), and network access via integrated APIs.
  • Sanitization: The agent is instructed to act on findings from these external sources without explicit sanitization, though the risk is significantly mitigated by mandatory human-in-the-loop approval gates before implementation plans and final merges.
  • [COMMAND_EXECUTION]: The skill makes extensive use of system CLI tools to manage the development environment.
  • Tools: Utilizes git for worktree and branch management, gt (Graphite) for managing stacked PRs, and gh (GitHub CLI) for PR submission, review tracking, and CI log inspection.
  • [EXTERNAL_DOWNLOADS]: The skill interacts with external services to synchronize state and fetch instructions.
  • Services: Communicates with GitHub and Linear via their respective CLI tools and MCP interfaces to manage the software development lifecycle.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 05:13 AM
Security Audit — agent-trust-hub — ship-autonomous