Use when you have a written implementation plan to execute in a separate session with review checkpoints
From zenflownpx claudepluginhub brewpirate/zen-flow --plugin zenflowThis skill uses the workspace's default tool permissions.
Searches prompts.chat for AI prompt templates by keyword or category, retrieves by ID with variable handling, and improves prompts via AI. Use for discovering or enhancing prompts.
Searches, retrieves, and installs Agent Skills from prompts.chat registry using MCP tools like search_skills and get_skill. Activates for finding skills, browsing catalogs, or extending Claude.
Compares coding agents like Claude Code and Aider on custom YAML-defined codebase tasks using git worktrees, measuring pass rate, cost, time, and consistency.
Load plan, review critically, execute all tasks, report when complete.
Announce at start: "I'm using the zenflow:exec-plan skill to implement this plan."
Note: This skill works best with subagent support. If subagents are available, consider using zenflow:dispatch instead for higher quality parallel execution.
TaskCreate for each section of the plan, then proceedCheck if the plan includes a Subagent Recommendation section (specifying which agents, skills, and parallelization to use).
For each task:
in_progress via TaskUpdatecompleted via TaskUpdate⚠️ A Stop hook will BLOCK you from finishing if you skip this step. You cannot complete a zenflow:exec-plan session without invoking zenflow:check-work.
After all tasks complete, do these in order:
skill: "zenflow:check-work". Do NOT manually run lint/format/test — the skill handles all 5 gates including the journal entry. Announce completion of each gate by name (Lint, Format, Tests, Docs, Journal) as you pass it. Skipping the journal gate is a rule violation.---
status: complete
completed: YYYY-MM-DD
---
STOP executing immediately when:
Ask for clarification rather than guessing.
Return to Review (Step 1) when:
Don't force through blockers - stop and ask.
zenflow:check-work before finishing. A hook enforces this.