agentleFS
Sign inStart free
hotcode-devhotcode-dev/zerofactory/skills/zerofactory-orchestration/SKILL.md

zerofactory-orchestration

Skill2 starsChanged 40 days agoApache-2.0View on GitHub

Set up specialized AI agent teams with role-specific toolsets, identities, and continuous agile workflows.

The file

We split this skill into the rules it gives an agent

4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Run hermes zerofactory setup to inspect or create the profiles.

    line 24
  2. Very specific:

    Run hermes zerofactory sync-profiles to synchronize system prompts from plugin templates.

    line 25
  3. Very specific:

    PR Creation & Continuous Review: When zf-builder completes the work, the dispatcher commits the branch, opens a GitHub Pull Request, and …

    line 35
  4. Very specific:

    Never commit directly to main: All agent development happens in isolated Git worktrees …

    line 39

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Multi-Agent Orchestration with Zero Factory

Zero Factory manages a team of specialized AI agents that operate in a continuous 24/7 agile loop, each with distinct identity, tools, and responsibilities.

Team Roles

Zero Factory automatically provisions and maintains the following specialist profiles in ~/.hermes/profiles/:

ProfileRoleIdentity / Responsibilities
zf-orchestratorPipeline OverseerMaster coordinator — manages the Kanban board, oversees goal decomposition, autonomously scans repositories to generate improvement TODO tasks, manages handoffs, and escalates blockers to human review.
zf-builderSenior Software EngineerWrites clean code and tests, operates inside automated Git worktrees, and ships features rapidly.
zf-reviewerQuality GatekeeperConducts thematic code reviews on GitHub Pull Requests (up to 3 rounds), verifying test adequacy, performance, and architecture.

Profile Management

Profiles are managed natively by the Zero Factory plugin (zerofactory):

  • Run hermes zerofactory setup to inspect or create the profiles.
  • Run hermes zerofactory sync-profiles to synchronize system prompts from plugin templates.
  • Profiles live in ~/.hermes/profiles/zf-*/ and inherit your default LLM settings.

Pipeline Architecture

The workflow is managed via the Zero Factory Kanban system with explicit Human-in-the-Loop (HITL) gates:

  1. Goal & Triage: User drops a high-level goal or epic into Triage. The kanban_decomposer (run by zf-orchestrator) automatically breaks the goal into child tasks.
  2. Actionable Backlog: Actionable child tasks and autonomous codebase improvement tasks filed by zf-orchestrator enter Todo.
  3. Autonomous Execution: The kanban dispatcher automatically provisions isolated Git worktrees, pulls/merges the latest default branch, and spawns zf-builder, moving tasks to Running.
  4. PR Creation & Continuous Review: When zf-builder completes the work, the dispatcher commits the branch, opens a GitHub Pull Request, and routes the ticket to zf-reviewer in Running. zf-reviewer runs multi-round review (up to 3 rounds) requesting changes or approving.
  5. Human Merge (HITL): Approved tasks move to Blocked for final human review and merge on GitHub. Once merged, the dispatcher marks the task as Done.

Best Practices

  • Never commit directly to main: All agent development happens in isolated Git worktrees (~/.zerofactory/workspaces/<board>/worktrees/<task_id>).
  • Use hermes zerofactory CLI: Inspect tasks, trigger manual dispatch cycles, and check queue health anytime.

Download now (ZIP)

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
What did you use it for? e.g. Reviewing pull requests in a Rails app

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.

Check your setup, free