Loop

Run chain

You might say: ticket to merge, the pipeline

A trigger starts a run. The run's change to the tracker or git host triggers the next run, through checks, automated review and human review, until a pull request merges. Each run hands its evidence to the next, such as screenshots of the bug or proof that the fix works. A new ticket starts the cycle again. The simplest chain is a loop in which each run picks the next task from a plan and commits it.

What it does
Joins short, stateless runs into one flow from ticket to merged code, with a gate or a human at each handoff.
Ask
Which change does each run make that starts the next one, and where can the chain stop?
Trouble
The chain stops silently between two runs, or one run's change starts two runs at once.

Loop: cycle

Trigger → Run → Automated checks → Automated review → Human review → Pull request ↺

Sources

  1. Anthropic, Building effective AI agents, Workflow: Prompt chainingPrompt chaining decomposes a task into a sequence of steps, where each LLM call processes the output of the previous one.
  2. GitHub, Events that trigger workflows, About events that trigger workflowsWorkflow triggers are events that cause a workflow to run.
  3. Martin Fowler, Continuous Integration, Every Push to Mainline Should Trigger a BuildEvery time the mainline receives a commit, the CI service checks out the head of the mainline into an integration environment and performs a full build.
  4. Martin Fowler, Continuous Delivery, Continuous DeliveryYou achieve continuous delivery by continuously integrating the software done by the development team, building executables, and running automated tests on those executables to detect problems.
  5. Jez Humble and David Farley, Continuous Delivery, Anatomy of the Deployment PipelineThe deployment pipeline is the key pattern that enables continuous delivery.
  6. GitHub, About protected branches, Require pull request reviews before mergingIf you enable required reviews, collaborators can only push changes to a protected branch via a pull request that is approved by the required number of reviewers
  7. Lauren Tan, pstack, Let loops start themselvesEvery stage hands over evidence. Repro attaches screenshots and video of the broken state, and the fix attaches before-and-after proof.
  8. Matt Pocock, Getting Started With Ralph, Getting Started With RalphRalph is a technique for running AI coding agents in a loop. You run the same prompt repeatedly. The AI picks its own tasks from a PRD. It commits after each feature.

Use it on your project

teardown-apply: hold "Run chain" (run-chain) from the teardown at https://teardownatlas.com/software-factory/ against my project

Paste it into a coding agent that has the teardown skills.

An AI agent wrote this map with the teardown skill. It can contain errors: check the sources before you rely on a part. About and legal