Android Static Analyzer

This skill reads and packages source code into LLM prompts and will cause your project source (including any hard-coded test credentials) to be sent to the agent's configured LLM when run. It also claims to upload results to a test platform but provides no upload code, API endpoint, or required credentials — verify how uploads are intended to occur before using. Recommendations: 1) Only run on non-sensitive or sanitized projects unless you trust the LLM endpoint and understand its data retention/privacy. 2) Confirm where the agent will call LLMs (cloud provider, on-prem) and whether you want that. 3) If you expect automatic reporting to a platform, require the skill to declare and use explicit platform_url and API token env vars and show upload code — otherwise treat the 'reporting' claim as unimplemented. 4) If you need to avoid sending source externally, run the included scripts locally and review their stdout (they print the LLM prompt) rather than letting the agent call an external model automatically. 5) Test on a small sample project first and inspect any files written (nav-graphs.json) to ensure no unexpected behavior.