Stats
Actions
Tags
From Agent Flow
Run a **repeatable checklist / to-do list** from a markdown file of `- [ ]` items — execute the open items, tick them `[x]`, and write the file back. Re-running resumes only the still-open items, so the file IS the durable, repeatable state. USE for a generic user to-do that should be runnable again and again: "run my TODO.md", "esegui la checklist", "go through CHECKLIST.md", "do the open items in release-steps.md". Each `- [ ]` line is itself the instruction for that item (optionally refined by --prompt). This is thin sugar over /agentflow:foreach with `--checkbox`: same engine, same durability and cross-turn resume, friendlier defaults. For "apply one operation to a JSON list / folder / glob" use /agentflow:foreach directly; to first GENERATE the list use /agentflow:enumerate; for one combined output use /agentflow:reduce.
How this skill is triggered — by the user, by Claude, or both
Slash command
/agentflow:checklist <checklist.md> [--prompt "<how to do each item>"] [--serial] [--model haiku|sonnet|opus] [--execution main-thread|subagent] [--dry-run]<checklist.md> [--prompt "<how to do each item>"] [--serial] [--model haiku|sonnet|opus] [--execution main-thread|subagent] [--dry-run]This skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
> **Make it visible:** in one line say you're starting a checklist run (file + run-id). `/agentflow:board`
Make it visible: in one line say you're starting a checklist run (file + run-id).
/agentflow:boardthen lists it on disk — the audit trail. All paths are relative to the workspace root.
You are running a repeatable checklist. This is /agentflow:foreach --checkbox with defaults — follow
the foreach skill's orchestration (${CLAUDE_PLUGIN_ROOT}/skills/foreach/SKILL.md); this file only
fixes the sugar.
TODO.md in the workspace
root if present. It must contain markdown task lines (- [ ] ..., completed ones - [x] ...).--prompt): default is "Complete this checklist task. The task text is in the
item's data. Do it end-to-end using the available tools, then report concisely what you did and any
follow-up. If the task is ambiguous or cannot be completed, say so and leave it for review." Override
with --prompt (e.g. --prompt "Draft the section described by each item").--serial → run items in list order, one at a time (sequential to-do). Default is parallel chunks.--model / --execution / --dry-run pass through to foreach (--dry-run = --validate-only +
report the open-item count without executing).checklist-<file-stem>-<date>):
node "${CLAUDE_PLUGIN_ROOT}/dist/state/foreach.js" init <run-id> --checkbox <file> --prompt "<operation>" [--serial]
--checkbox pre-marks existing [x] lines as done, so only open items are scheduled (this is what
makes re-runs resume cleanly). For --dry-run, add --validate-only and stop here, reporting
total/pending.claim a chunk → dispatch each item to a
subagent (or do it inline for main-thread) with the operation + the item's task text → record results
with complete/complete-batch. The state file is the only source of truth; subagents never write
state.node "${CLAUDE_PLUGIN_ROOT}/dist/state/foreach.js" view <run-id> --checkbox <file>
- [x]; failed/blocked ones stay - [ ] and are reported for the next run./agentflow:checklist <file> later picks up exactly the still-open items.--stop-file; to cap cost, pass --max-usd (see foreach).npx claudepluginhub alesaiani/agentflow --plugin agentflowGuides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.