deploy

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of the Bash tool to execute development commands including gh (GitHub CLI), git, docker, and ssh for managing workflows and deployments.
  • [EXTERNAL_DOWNLOADS]: Fetches workflow statuses, run logs, and repository metadata from the official GitHub API via the gh CLI. It also performs artifact verification against PyPI and Docker registries during the release process.
  • [SAFE]: Implements a comprehensive safety classification system (READ, CREATE, MODIFY, SERVICE, DELETE, PRIVILEGE) that requires explicit user confirmation via the AskUserQuestion tool before executing any commands that modify state or trigger actions.
  • [COMMAND_EXECUTION]: Probes for and executes local repository scripts (e.g., version bumping or post-release scripts) to integrate with the project's existing release tooling, but only after validating the environment.
  • [SAFE]: Automatically manages project-specific deployment configuration in CLAUDE.local.md and ensures this file is added to .gitignore to prevent sensitive local data from being committed.
  • [SAFE]: Generates a companion agent (deploy-admin.md) using a local template and the project's specific configuration, adhering to least-privilege principles and maintaining the same safety gates as the main skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 06:23 PM
Security Audit — agent-trust-hub — deploy