The Agent Skills Directory

[COMMAND_EXECUTION]: The skill generates a POSIX shell script and writes it to ~/.claude/statusline-command.sh. It applies execution permissions using chmod +x and runs the script against a test JSON payload to verify the output during the configuration workflow.
[PROMPT_INJECTION]: The skill processes untrusted content by reading the existing ~/.claude/statusline-command.sh file before modification. Ingestion points: Data is read from the existing script file during the custom request workflow. Boundary markers: No delimiters or 'ignore instructions' markers are used when reading the script file. Capability inventory: The skill has the ability to write files, change file permissions, and execute shell commands. Sanitization: No validation or sanitization is performed on the content of the existing script file before it is incorporated into the agent's context for generating a new version.

statusline-config