agentleFS
lza6lza6/claw-code-tingfeng/skills/pipeline_orchestrator/SKILL.md

pipeline_orchestrator

Skill5 starsChanged 13 days agoMITView on GitHub

Configurable pipeline orchestrator for sequencing RALPLAN -> team-exec -> ralph

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is 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.

  • 1

    Very specific

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

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

    Every stage implements the PipelineStage interface:

    line 32

≈ “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

Pipeline Orchestrator Skill

$pipeline_orchestrator provides configurable pipeline orchestration for Clawd Code.

It sequences stages through a uniform PipelineStage interface, with state persistence and resume support.

Default Pipeline

The canonical pipeline sequences:

RALPLAN (consensus planning) -> team-exec (agent swarm) -> ralph (verification)

Configuration

Pipeline parameters are configurable per run:

ParameterDefaultDescription
max_ralph_iterations10Ralph verification iteration ceiling
worker_count2Number of agent workers
agent_typeexecutorAgent type for workers

Stage Interface

Every stage implements the PipelineStage interface:

class PipelineStage(Protocol):
    name: str
    async def run(ctx: StageContext) -> StageResult: ...
    def can_skip(ctx: StageContext) -> bool: ...

Stages receive a StageContext with accumulated artifacts from prior stages and return a StageResult with status, artifacts, and duration.

Built-in Stages

  • ralplan: Consensus planning phase. Generates structured task breakdown, dependencies, and risks using the planner agent. Skips only when both prd-*.md and test-spec-*.md planning artifacts already exist.
  • team-exec: Parallel agent swarm execution using team mode. Always the execution backend for Clawd Code.
  • ralph-verify: Iterative verification loop using the ralph agent with configurable iteration count.

State Management

Pipeline state persists in .claude/workflow/pipeline_state.json (compatible with GoalX durable surfaces).

  • On start: state_write({mode: "pipeline", active: True, current_phase: "stage:ralplan"})
  • On stage transitions: state_write({mode: "pipeline", current_phase: "stage:<name>"})
  • On completion: state_write({mode: "pipeline", active: False, current_phase: "complete"})

The HUD dashboard renders pipeline phase automatically. Resume is supported from the last incomplete stage.

API

from src.workflow.pipeline_orchestrator import (
    run_pipeline,
    create_autopilot_pipeline_config,
    create_ralplan_stage,
    create_team_exec_stage,
    create_ralph_verify_stage,
)

config = create_autopilot_pipeline_config(
    task="Build feature X",
    stages=[
        create_ralplan_stage(),
        create_team_exec_stage(worker_count=3, agent_type="executor"),
        create_ralph_verify_stage(max_iterations=15),
    ]
)

result = await run_pipeline(config)

Relationship to Other Modes

  • team: Pipeline delegates execution to team mode (agent swarm)
  • ralph: Pipeline delegates verification to ralph mode (configurable iterations)
  • ralplan: Pipeline's first stage runs RALPLAN consensus planning

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.