The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses shell commands to interact with the version control system. Specifically, it executes git add and git commit to finalize changes made during the archiving and promotion process.
[PROMPT_INJECTION]: The skill processes markdown files that may have been generated by other tasks or agents, which presents a surface for indirect prompt injection if those files contain malicious instructions.
Ingestion points: Reads markdown files located in skills/aitc-task-<batch>/ and existing skills in skills/ or ~/.claude/skills/.
Boundary markers: No explicit delimiters or 'ignore' instructions are used when reading and interpolating these files.
Capability inventory: Performs file reading, writing, and deletion; directory removal; and shell command execution (git).
Sanitization: The skill incorporates human-in-the-loop verification by requiring the user to review a summary table and confirm diffs before any promotion, deletion, or merge operation is executed.
[SAFE]: The identified behaviors are consistent with the skill's stated purpose. Access to the agent's global configuration directory and the use of version control commands are necessary for lifecycle management, and the design includes significant safeguards via user interaction.

aitc-workflow-lifecycle