overnight-pipeline

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied feature descriptions to generate automated workflows. Ingestion points: Command line arguments and codebase content. Boundary markers: None explicitly defined. Capability inventory: Shell script execution, file-system writes, and git operations. Sanitization: Mitigated by mandatory user review before execution.
  • [DYNAMIC_EXECUTION]: The skill dynamically generates shell scripts based on user input and selected pipeline patterns. These scripts follow standard templates for software development automation.
  • [COMMAND_EXECUTION]: The skill generates commands for local tools like git, bun, and the claude CLI. These actions are performed within the local environment and are subject to user verification.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 10:29 AM
Security Audit — agent-trust-hub — overnight-pipeline