mobilebuildmcp-test-boundary-review
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from local test files and external documentation, which could contain instructions designed to manipulate the agent's behavior.
- Ingestion points: Local files (
src/**,package.json) and external documentation URLs (xcodebuildmcp.com). - Boundary markers: Absent; there are no instructions to differentiate between data and instructions in the inspected files.
- Capability inventory: The agent is instructed to execute shell commands such as
npm test,npm run typecheck, andnpx skill-check. - Sanitization: Absent.
- [COMMAND_EXECUTION]: The skill instructs the agent to run validation commands like
npm testandnpx skill-checkin the local environment. - [EXTERNAL_DOWNLOADS]: The skill references and inspects documentation content from the external domain
xcodebuildmcp.com.
Audit Metadata