The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local Python script (journal_append.py) located in a path defined by the ${LEYLINE} environment variable. This script is used to automate the formatting and appending of decision logs to local documentation files (docs/tradeoffs.md and docs/lessons-learned.md).
[SAFE]: The skill implements a strong human-in-the-loop security control by explicitly instructing the agent to draft the entry and then 'Show the draft to the human and let them confirm or edit' before execution. It further mandates that the agent 'Do not auto-write without the confirm step,' preventing autonomous or hidden file modifications.

decision-journal