The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute a local bash script (zmm-mvp/scripts/top_posts.sh) to perform voice calibration by analyzing the user's historical post data stored in a CSV file.
[COMMAND_EXECUTION]: The agent is directed to use the pbcopy command to automatically copy the final generated content to the system clipboard after writing it to a file.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest arbitrary user-provided content such as "fragments," chat screenshots, or notes as input for generating posts.
Ingestion points: Untrusted user input enters the agent context in Phase 0 of the workflow.
Boundary markers: The instructions do not define specific delimiters or "ignore instructions" wrappers for the ingested fragments.
Capability inventory: The skill possesses capabilities for local file system writes, script execution, and clipboard access.
Sanitization: The skill lacks explicit sanitization or filtering logic for the content of ingested fragments prior to processing.
[DYNAMIC_EXECUTION]: The skill utilizes dynamic execution by invoking a shell script at runtime to locate and parse data files within the user's workspace.

zmm-mvp