2.6k
GitHub Stars
2
Bundled Files
3 weeks ago
Catalog Refreshed
1 month ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstart where the catalogue uses aiagentskills.
npx veilstart add skill openclaw/skills --skill team-task-dispatch- _meta.json290 B
- SKILL.md5.2 KB
Overview
This skill coordinates team task execution on OpenAnt by providing structured commands and a clear workflow for creating, claiming, submitting, and reviewing subtasks. It assumes a team has accepted a parent task and splits work into manageable subtasks with roles for LEAD and WORKER. Use the CLI with JSON output for automation and agent polling.
How this skill works
The skill guides team members through an inbox-driven workflow: check the inbox, list subtasks, claim open work, mark progress, submit deliverables, and let the LEAD review and verify. LEADs create and organize subtasks (priority, dependencies, deadlines) while WORKERs claim, start, and submit subtasks. All commands are run via the OpenAnt CLI with --json for deterministic, machine-parseable responses suitable for autonomous agents.
When to use it
- Your team has accepted a parent task and work must be divided into subtasks.
- You need to coordinate claiming and assignment of granular work items.
- You want automated agents to poll for new work and act without manual parsing.
- A LEAD must review and verify completed subtasks before final submission.
- You need progress visibility and dependency enforcement across subtasks.
Best practices
- Always append --json to CLI commands for structured output and automation.
- LEADs break tasks into clear, dependency-aware subtasks with deadlines and priorities.
- Workers only claim OPEN subtasks and ensure dependencies are VERIFIED first.
- Poll the inbox periodically and follow decision logic: claim, work, submit, or review.
- Use concise submission notes (links to PRs, screenshots) to make reviews faster.
Example use cases
- A LEAD splits a feature into API design, backend implementation, and tests with dependencies.
- A worker polls the inbox, claims an OPEN subtask, marks it in-progress, and submits a PR link.
- A LEAD reviews submitted subtasks, approves or rejects with comments, and reassigns if needed.
- An autonomous agent checks inbox every few minutes and picks subtasks matching its capabilities.
- Track overall task progress and submit the parent task once all subtasks are verified.
FAQ
Yes. --json produces structured responses necessary for reliable automation and parsing.
Who can create subtasks?
Only the LEAD for the accepted parent task can create subtasks; others will see an error.
Why can't I claim a subtask?
Common causes: the subtask is not OPEN, you are not a task participant, or a dependency is not VERIFIED.