Skip to content
Agentmart
Submit a tool

cmd-done-check

Ask the agent whether it finished everything or has more to do — a lightweight completeness gate for the end of any task

Skill by Olshansk

IndexedMIT★ 9
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/olshansk-cmd-done-check

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Ask the agent whether it finished everything or has more to do — a lightweight completeness gate for the end of any task

Use cases: productivity, agent-building

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/cmd-done-check

What's inside

Skill name
cmd-done-check
Description
Ask the agent whether it finished everything or has more to do — a lightweight completeness gate for the end of any task
License
MIT
Instructions
42 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Done Check

A quick self-interrogation to catch loose ends before declaring a task complete.

## Instructions

Answer each question honestly. If anything is incomplete, say so and ask the user whether to continue.

### 1. Did you do everything the user asked?

- Re-read the original request. Is every explicit ask addressed?
- Did you skip or defer anything without telling the user?
- Did you say "I'll also..." or "next, we should..." and then not do it?

### 2. Did you leave any loose ends?

- Are there TODOs, FIXMEs, or placeholder values you introduced that should be resolved now?
- Are there commented-out blocks, debug prints, or temporary stubs in the output?
- Are there follow-up steps you mentioned but didn't take?

### 3. Did you handle side effects?

- Are there related files, configs, or tests that should be updated but weren't?
- Did any rename, move, or deletion leave dangling references?
- If you added a skill/command/export — is it wired up and discoverable?

### 4. Are docs and metadata consistent?

- If you added or changed something user-facing, is the README or AGENTS.md updated?
- Are variable names, comments, and doc strings consistent with what shipped?

## Output

**If everything is done:** say "Done — nothing left." in one line.

**If something is incomplete:** list each item as a bullet with what's missing and why, then ask:

> Want me to continue and handle these now, or are you good?

Do not pad the response. One sentence per finding.

From Olshansk/agent-skills, MIT.

More from Olshansk/agent-skills

Signals

Stars
9
Forks
0
Open issues
0
Last commit (this folder)
May 20, 2026 (5 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from Olshansk

Cite this page

cmd-done-check skill. Agent Mart. https://agentmart.dev/skills/olshansk-cmd-done-check (last updated 2026-10-24)