do-subtasks
Skill3 starsChanged 36 days agoMITView on GitHub
Executes exactly one approved subtask at a time, then returns a conventional commit message suggestion for human review and commit. Use when the user wants strict one-task-at-a-time implementation with manual commit control and explicit "next task" prompts.
In this repository9 other files it gives its agents.
- AGENTS.mdAGENTS.md
- audit-project-rulesSkill
- changelogSkill
- pest-arch-reviewSkill
- qaSkill
- retroSkill
- simplify-agent-instructionsSkill
- to-prSkill
- to-subtasksSkill
The file
We split this skill into the rules it gives an agent
9 rules. About 0 are common, 8 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 11
Do not begin additional subtasks in the same response.
- Distinctive: line 12
When that subtask is complete, stop implementation and return only a conventional commit message suggestion.
- Distinctive: line 15
Wait for the human to tell you to move to the next task.
- Distinctive: line 19
When the human tells you to continue, begin work immediately.
- Distinctive: line 20
Do not ask for confirmation.
- Distinctive: line 21
Do not restate or confirm which task you are starting.
- Distinctive: line 22
If the human asks for the next subtask and there are no remaining agreed-upon subtasks, do not implement anything.
- Very specific: line 29
No scope section like feat(api): ...
- Distinctive: line 30
No body text, bullets, or extra commentary.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Do Subtasks
Workflow
- Implement exactly one subtask at a time.
- Do not begin additional subtasks in the same response.
- When that subtask is complete, stop implementation and return only a conventional commit message suggestion.
- The commit message must have no scope and no body.
- The human reviews and commits; do not commit automatically.
- Wait for the human to tell you to move to the next task.
Next-task behavior
- When the human tells you to continue, begin work immediately.
- Do not ask for confirmation.
- Do not restate or confirm which task you are starting.
- If the human asks for the next subtask and there are no remaining agreed-upon subtasks, do not implement anything.
- In that case, tell the human that all agreed-upon subtasks have already been implemented.
Output Rules
- At subtask completion, output a single-line conventional commit message only.
- Allowed format:
<type>: <short description> - No scope section like
feat(api): ... - No body text, bullets, or extra commentary.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.