agentleFS
Sign inStart free
elfrostelfrost/ai-patchlab/.agents/skills/pipeline/SKILL.md

pipeline

Skill1 starsChanged 31 days agoMITView on GitHub

Execute a predefined or custom pipeline workflow defined in `.claude/pipelines/*.yml`. Codex walks the steps inline since it has no subagent primitive. Use when the user wants to run a sequenced workflow like `pipeline feature` or `pipeline release`.

29 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use follow-up, resume and 6 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • n24q02mSkill

    follow-up

    Find emails needing follow-up -- unanswered sent emails, flagged unreplied, draft contextual responses

    Also in 3 projects that use this

  • genomewalkerSkill

    resume

    Resume a thread by loading its ~800-token context capsule

    Also in 3 projects that use this

  • genomewalkerSkill

    explore

    RLM-style recursive memory exploration - dynamically navigate the memory graph

    Also in 3 projects that use this

  • AnthropicSkill

    analyze

    Answer data questions -- from quick lookups to full analyses. Use when looking up a single metric, investigating what's driving a trend or drop, comparing segments over time, or preparing a formal data report for stakeholders.

    Also in 3 projects that use this

  • dubwoahSkill

    ask

    Answer questions about the codebase without generating or modifying any code. Use when the user wants to understand how something works, find where something is defined, or explore the project structure. Invoked with /ask.

    Also in 3 projects that use this

  • ibahgatSkill

    autopilot

    Run autonomous multi-stage execution loops until acceptance criteria pass or blockers are reached.

    Also in 3 projects that use this

The file

We split this skill into the rules it gives an agent

7 rules. 0 are common, 5 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 5

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

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

    Look for the pipeline file in: .claude/pipelines/<name>.yml then .yaml

    line 12
  2. Distinctive:

    If not found, list available pipelines and ask the user to choose

    line 13
  3. Very specific:

    Check the mode field — if it requires project mode, verify .ezproject.json#mode matches.

    line 15
  4. Distinctive:

    Always show the execution plan and get approval before running

    line 36
  5. Distinctive:

    Always respect dependency order

    line 37
  6. Distinctive:

    If a gate is rejected, stop gracefully

    line 38
  7. Distinctive:

    Do not skip ahead to dependent steps if a predecessor failed

    line 39

“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “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

Execute a YAML-defined workflow step by step. Each type: agent step is executed by reading the corresponding agent's prompt at template/.claude/agents/<agent>.md and following it inline.

Phase 1: Load

  1. Parse $ARGUMENTS for pipeline name (first word) and input (rest)
  2. Look for the pipeline file in: .claude/pipelines/<name>.yml then .yaml
  3. If not found, list available pipelines and ask the user to choose
  4. Parse the YAML
  5. Check the mode field — if it requires project mode, verify .ezproject.json#mode matches; warn if not

Phase 2: Show plan

Display the steps in order with a table (id, type, agent, depends_on). Highlight gates (human approval) and ask for confirmation before executing.

Phase 3: Execute

Walk steps in dependency order. For each step:

  • type: agent — read .claude/agents/<agent>.md (or project-mode equivalent at project-mode/.claude/agents/), follow that agent's process inline using its prompt. Substitute {{step_id.output}} with previous step outputs and {{input}} with the user's input. Capture the output.
  • type: command — invoke the corresponding skill described in .agents/skills/<name>/SKILL.md.
  • type: validation — run each command in commands list. Retry up to retry count if any fail. If still failing, STOP and ask the user.
  • type: gate — show the message and ask the user. Stop gracefully on rejection.
  • type: summary — collect previous step outputs and produce a structured summary.

Phase 4: Report

Produce a final report:

  • Steps table with status per step
  • Summary
  • Suggested next actions

Rules

  • Always show the execution plan and get approval before running
  • Always respect dependency order
  • If a gate is rejected, stop gracefully
  • Do not skip ahead to dependent steps if a predecessor failed
  • Without subagent isolation, run agent steps sequentially even if Claude would parallelize them

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