The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses Git commands (add, commit, push) and searching tools (grep, awk) to update and query the development log. These commands are executed as part of the primary logging workflow.
[INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from existing log files and user conversations. 1. Ingestion points: Reads DEVLOG.md files and user input during sessions. 2. Boundary markers: Relies on Markdown headers and structured entry sections. 3. Capability inventory: Performs file writes to the repository and user home directory, and network operations via Git push. 4. Sanitization: Instructs the agent to proactively redact secrets and sensitive data.

tl-devlog