The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local commands such as git and grep through a bundled bash script (detect.sh) to perform static analysis of the repository. These operations are restricted to the local filesystem and follow secure coding practices, including proper variable quoting and error handling.\n- [DATA_EXFILTRATION]: All analysis is performed locally, and results are returned to the agent's context. No network operations, external requests, or data exfiltration patterns were detected.\n- [PROMPT_INJECTION]: The skill's instructions and action definitions are technically focused and do not contain patterns intended to bypass safety filters, extract system prompts, or override agent constraints.\n- [REMOTE_CODE_EXECUTION]: The skill utilizes its own distributed scripts for auditing, which is an expected and safe behavior for a local utility. It does not download or execute code from external or untrusted sources.

audit-encapsulation