The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides a script (scripts/inspect-project.py) and instructs the agent to execute it locally to analyze an Android project. This is a core part of its diagnostic functionality.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local source files via the inspection script. While the output could be influenced by malicious code within a project being analyzed, the script uses safe regex-based pattern matching and does not execute the analyzed code. The skill also requires the user to follow an evidence-based reasoning process that mitigates the risk of blindly following instructions found in data.
[SAFE]: The skill includes explicit instructions to redact sensitive information such as MAC addresses and names, promoting privacy. It operates entirely on local project files and lacks any network exfiltration or remote code execution capabilities.

android-ble-gatt-status