mobilebuildmcp-docs-release-review
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from project files such as
README.mdandCHANGELOG.mdwhich could contain malicious instructions designed to influence agent behavior during review. - Ingestion points: Accesses
README.md,CHANGELOG.md,package.json, and website manifest files. - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt interpolation logic.
- Capability inventory: Executes subprocesses via
npm run build,node scripts/generate-github-release-notes.mjs, andnpx skill-check. - Sanitization: No evidence of sanitization or validation of the content within the ingested files before processing.
- [COMMAND_EXECUTION]: The skill instructs the agent to execute local commands and scripts (
npm run build,scripts/release.sh, andscripts/generate-github-release-notes.mjs) to validate documentation and release consistency. While these are project-internal scripts, their execution on local environments carries inherent risk if the scripts themselves are modified by an attacker.
Audit Metadata