xcodebuildmcp-packaging-resource-review

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill instructions are focused on legitimate build and packaging review tasks. No signs of obfuscation, malicious command patterns, or unauthorized data access were found.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute standard project build and validation commands, including npm run build, npm run verify:portable, and npx skill-check. These are expected development activities performed within the local project context.
  • [PROMPT_INJECTION]: The skill involves reviewing external project files (e.g., package.json, scripts, manifests), which introduces a potential surface for indirect prompt injection. This is inherent to code review tasks and is managed by the agent's internal safety guardrails.
  • Ingestion points: Files specified in the 'Files to inspect' list (e.g., package.json, scripts/copy-build-assets.js, manifests/**).
  • Boundary markers: None explicitly mentioned in the instructions.
  • Capability inventory: Execution of local build scripts and validation tools via npm and npx.
  • Sanitization: None specified for the content of the reviewed files.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 03:42 AM
Security Audit — agent-trust-hub — xcodebuildmcp-packaging-resource-review