The Agent Skills Directory

[COMMAND_EXECUTION]: The skill retrieves and executes shell commands (e.g., for linting and testing) that are defined in external context notes like [[Context {Project} Tests]]. This behavior allows for dynamic execution of commands sourced from externally modifiable files.
[CREDENTIALS_UNSAFE]: The 'Secrets scan' battery identifies sensitive information such as API keys and tokens in the code diff. The skill explicitly instructs the agent to copy discovered secrets 'verbatim' into a markdown report file, which results in sensitive credentials being stored in plain text outside of secure configurations.

conducty-ship