agentsclimarketplace

Zeoel dispatch

Skill goharabbas321/zeoel/skills/zeoel-dispatch

Zeoel-AI is a production-grade, multi-agent SaaS software engineering agency. Instead of single, fragile LLM prompts that lose context, Zeoel-AI coordinates a complete team of 33 specialized agents (UX designers, database architects, systems engineers, security auditors, QA engineers, etc.) to brainstorm, plan, build, and verify entire software.

Install
npx -y skills add goharabbas321/zeoel --skill zeoel-dispatch

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Guidelines for Phase 3 Execution. Dispatches sub-agents for specialized tasks, enforces strict Red-Green-Refactor TDD, updates progress files, and handles recovery protocols.

SKILL.md

2.2 KB, as published. Nobody here has run it

Zeoel Dispatch & Execution Skill (Phase 3)

This skill governs Phase 3: Execution.

Red-Green-Refactor TDD Workflow

Sub-agents must strictly follow Test-Driven Development (TDD) for all tasks containing application code:

  1. Red Phase: Write the failing test first, referencing the specifications.
  2. Green Phase: Write the minimal application code in frontend/ or backend/ to make the test pass.
  3. Refactor Phase: Clean up, deduplicate, and document the code.

Post-Task Checkpoint

After completing a task:

  1. Update docs/sprint-N/progress.md with:

    • Status: ✅ Done
    • Test files: list of created test files
    • Timestamp

    [!IMPORTANT] Strict Task Matching Guard: When editing progress.md, you MUST target only the specific row matching the current task's unique ID/Number (e.g., Task K). NEVER use a global or generic search-and-replace on the agent's name (e.g. matching all lines containing zara or mahdi), as this will accidentally mark future or combined tasks (e.g. zara + mahdi) as completed prematurely. Ensure only the single, specific task row is modified.

  2. Run /graphify . --update to update the codebase semantic relationship graph.

  3. Update docs/sprint-N/deferred.md if any scope was deferred.

  4. Update relevant sections of PROJECT_BRIEF.md directly (such as Database Schema, API Specification, UI Layout Map) to reflect the exact implementation details.

  5. Append a message describing the completed task, code note findings, new API routes/endpoints, database schema changes, and important design details to the ## Centralized Agent Messaging & Findings Log section at the end of PROJECT_BRIEF.md. This log serves as the shared message board for the CEO and sub-agents. Format your entry as follows:

    #### [Agent Name/Role] Findings - Task K [Timestamp]
- **Task**: [Task name / description]
- **Changes/Findings**: [Brief summary of the changes and findings]
- **Dependency/Messaging Alert**: [Important notes for next tasks or dependent agents]

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.