The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the Bash tool to execute a local maintenance script, larch.sh, located within the ${CLAUDE_PLUGIN_ROOT} directory. This script removes stale temporary files from ~/.cache/larch/sessions/, /tmp, and the $TMPDIR path.
[INDIRECT_PROMPT_INJECTION]: The skill parses standard output from the maintenance script to extract removal counts and relay them to the user. Ingestion points: Data enters the agent's context through the parsed stdout of the larch.sh command execution in SKILL.md. Boundary markers: There are no delimiters or instructions used to distinguish the script's output from the agent's logic. Capability inventory: The skill has the capability to execute shell commands and write to the user-facing output. Sanitization: The skill looks for specific keys (e.g., SESSION_COUNT) but does not sanitize the values parsed from these keys before reporting them.

cleanup