Plan Do Check Act

PDCA workflow automation with session binding and progress recovery. Requires Python 3.6 and Git (optional).

This skill appears coherent for PDCA automation, but take these precautions before installing or running it: - Use a dedicated workspace directory (the default is ~/pdca-workspace) and do not point the tool at system, home-root, or project root directories containing sensitive files. The scripts recursively scan the workspace and report on recently modified filenames and git status. - Review archive/cleanup behavior and always run cleanup with --dry-run first to confirm what will be deleted. cleanup() will remove archived .md files older than the configured threshold. - The skill uses subprocess calls to run Git (git status, git init). If you opt into git initialization, be aware it will create a .git repository under the chosen workspace. - The skill asks for interactive confirmations in some flows (archive prompt). If you plan to let an agent invoke it autonomously, consider whether you want to permit non-interactive runs that could archive/delete files; avoid granting autonomous runs if you are uncomfortable with that. - If you want higher assurance, inspect or run the bundled scripts locally in a safe test workspace before allowing the skill to operate on real project data. If you want, I can highlight exact lines that perform file deletion, git operations, and workspace traversal so you can review them quickly.

Type: OpenClaw Skill Name: plan-do-check-act Version: 1.2.0 The skill implements a legitimate Plan-Do-Check-Act (PDCA) workflow for task management and progress tracking. It provides scripts (manage_plan.py, check_progress.py) to create checklists, monitor file modifications/Git status for activity, and archive completed tasks. The code uses standard Python libraries, lacks network exfiltration capabilities, and limits file operations to a user-defined workspace. While it contains a hardcoded Discord ID (1488795270769676318) in documentation and help strings, it is used strictly as a placeholder example for session binding logic.