skills/reason-machines/trending-skills/claude-skill-app-onboarding-questionnaire/Gen Agent Trust Hub
claude-skill-app-onboarding-questionnaire
Warn
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation instructs the user to clone and install code from a third-party GitHub repository (github.com/adamlyttleapps/claude-skill-app-onboarding-questionnaire.git) which is not a verified or trusted organization.
- [DATA_EXFILTRATION]: The skill is designed to automatically read and analyze sensitive project files, specifically
Info.pliston iOS andAndroidManifest.xmlon Android, to extract permission usage descriptions and app metadata. - [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted local source code and configuration files to drive its code generation engine, which lacks explicit sanitization or boundary markers. * Ingestion points: App source code, manifest files, and property lists (as described in SKILL.md). * Boundary markers: None mentioned for the ingested content. * Capability inventory: The skill can read project files and write new source code files to the project directory. * Sanitization: No sanitization or validation process for the ingested code is described.
Audit Metadata