The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes local system commands to determine the project root and execute its core logic.\n
Evidence: SKILL.md instructs the agent to run git rev-parse --show-toplevel and uv run python scripts/archive_todo.py.\n
Evidence: scripts/archive_todo.py uses subprocess.run() with a fixed list of arguments to resolve the git root.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes user-controlled data from TODO.md which could theoretically contain instructions.\n
Ingestion points: scripts/archive_todo.py reads the contents of TODO.md from the repository root.\n
Boundary markers: Absent; the file content is processed as raw text.\n
Capability inventory: The skill has access to the local filesystem (read/write to TODO.md and .ai/) and can execute local git commands via subprocess.run.\n
Sanitization: The archival logic uses specific regex patterns to identify tasks and headings, ignoring other content.\n- [SAFE]: No network exfiltration, hardcoded credentials, or obfuscated code patterns were found. File operations are confined to the repository scope and follow expected behavior for a task management tool.

todo-archive