Title: Flow-Next · Agents generate. Flow-Next proves.
Open Graph Title: Flow-Next · Agents generate. Flow-Next proves.
X Title: Flow-Next · Agents generate. Flow-Next proves.
Description: Flow-Next is a workflow plugin for your coding agent. It takes an idea, a ticket, a bug report or a slow page all the way to a pull request in about the time plain Claude Code, or your harness, takes, often less, and large features in about half the time. The result is better than the plain agent's even before any review; Flow adds review by another model and QA in the real app where the risk calls for them, widening the gap to up to +48% better outcomes on large and long-horizon work, and every done comes with proof. One routing block sends each job to the model or harness that fits it. Run it unattended and it records every decision in the PR.
Open Graph Description: Flow-Next is a workflow plugin for your coding agent. It takes an idea, a ticket, a bug report or a slow page all the way to a pull request in about the time plain Claude Code, or your harness, takes, often less, and large features in about half the time. The result is better than the plain agent's even before any review; Flow adds review by another model and QA in the real app where the risk calls for them, widening the gap to up to +48% better outcomes on large and long-horizon work, and every done comes with proof. One routing block sends each job to the model or harness that fits it. Run it unattended and it records every decision in the PR.
X Description: Flow-Next is a workflow plugin for your coding agent. It takes an idea, a ticket, a bug report or a slow page all the way to a pull request in about the time plain Claude Code, or your harness, takes, often less, and large features in about half the time. The result is better than the plain agent's even before any review; Flow adds review by another model and QA in the real app where the risk calls for them, widening the gap to up to +48% better outcomes on large and long-horizon work, and every done comes with proof. One routing block sends each job to the model or harness that fits it. Run it unattended and it records every decision in the PR.
Opengraph URL: https://flow-next.dev/
X: @gmickel
Domain: flow-next.dev
{"@context":"https://schema.org","@type":"SoftwareApplication","name":"Flow-Next","description":"Flow-Next is a workflow plugin for your coding agent. It takes an idea, a ticket, a bug report or a slow page all the way to a pull request in about the time plain Claude Code, or your harness, takes, often less, and large features in about half the time. The result is better than the plain agent's even before any review; Flow adds review by another model and QA in the real app where the risk calls for them, widening the gap to up to +48% better outcomes on large and long-horizon work, and every done comes with proof. One routing block sends each job to the model or harness that fits it. Run it unattended and it records every decision in the PR.","url":"https://flow-next.dev/","applicationCategory":"DeveloperApplication","operatingSystem":"Linux, macOS, Windows","softwareVersion":"9.17.0","license":"https://github.com/gmickel/flow-next/blob/main/LICENSE","downloadUrl":"https://flow-next.dev/install/","softwareRequirements":"A supported coding agent, Python 3.11+, jq, gh, and your project tools"}
| theme-color | #041116 |
| og:site_name | Flow-Next |
| og:type | website |
| og:image | https://flow-next.dev/og-image.png |
| og:image:width | 1200 |
| og:image:height | 630 |
| og:image:alt | The Flow-Next mark beside a review receipt stamped SHIP. |
| twitter:card | summary_large_image |
| twitter:creator | @gmickel |
| twitter:image | https://flow-next.dev/og-image.png |
Links:
| Flow-Next | https://flow-next.dev/ |
| Flow-Next Docs 9.17.0 | https://flow-next.dev/ |
| ◌ Introduction | https://flow-next.dev/introduction/ |
| ◌ Install | https://flow-next.dev/install/ |
| ◌ First 30 Minutes | https://flow-next.dev/first-30-minutes/ |
| ◌ How Flow chooses | https://flow-next.dev/choosing-your-route/ |
| ◌ One change through review | https://flow-next.dev/guides/worked-example/ |
| ◌ For agents | https://flow-next.dev/for-agents/ |
| ◌ Overview | https://flow-next.dev/understand/ |
| ◌ Why Flow-Next | https://flow-next.dev/understand/why-flow-next/ |
| ◌ The Pipeline | https://flow-next.dev/understand/pipeline/ |
| ◌ How work gets proven | https://flow-next.dev/understand/how-work-gets-proven/ |
| ◌ Compose the pipeline | https://flow-next.dev/understand/compose-the-pipeline/ |
| ◌ Flow and the road ahead | https://flow-next.dev/understand/flow-and-the-road-ahead/ |
| ◌ Explore first, then capture | https://flow-next.dev/understand/explore-first/ |
| ◌ How it compounds | https://flow-next.dev/understand/how-it-compounds/ |
| ◌ Architecture & harnesses | https://flow-next.dev/understand/architecture/ |
| ◌ Concepts | https://flow-next.dev/understand/concepts/ |
| ◌ Overview | https://flow-next.dev/guides/ |
| ◌ Writing specs | https://flow-next.dev/guides/writing-specs/ |
| ◌ From prototype to spec | https://flow-next.dev/guides/prototype-to-spec/ |
| ◌ Customizing the spec scaffold | https://flow-next.dev/guides/spec-scaffold/ |
| ◌ Tasks, dependencies & parallelism | https://flow-next.dev/guides/tasks/ |
| ◌ Review workflow | https://flow-next.dev/guides/review-workflow/ |
| ◌ Orchestration & routing | https://flow-next.dev/guides/model-routing/ |
| ◌ Optional Jev judgments | https://flow-next.dev/guides/jev-judge/ |
| ◌ Discovery before capture | https://flow-next.dev/guides/discovery/ |
| ◌ Live-app QA & driving UIs | https://flow-next.dev/guides/live-qa/ |
| ◌ Keep the feature map current | https://flow-next.dev/guides/keep-feature-map-current/ |
| ◌ For teams | https://flow-next.dev/guides/for-teams/ |
| ◌ Regulated environments | https://flow-next.dev/guides/regulated-environments/ |
| ◌ Cookbook | https://flow-next.dev/guides/cookbook/ |
| ◌ Going Autonomous | https://flow-next.dev/autonomy/going-autonomous/ |
| ◌ Flow --auto - the build loop | https://flow-next.dev/autonomy/pilot/ |
| ◌ Goal mode - toward an outcome | https://flow-next.dev/autonomy/goal-mode/ |
| ◌ Land - the ship loop | https://flow-next.dev/autonomy/land/ |
| ◌ Driving a loop | https://flow-next.dev/autonomy/driving-a-loop/ |
| ◌ Unattended operation | https://flow-next.dev/autonomy/unattended-operation/ |
| ◌ Tracker sync | https://flow-next.dev/integrations/tracker-sync/ |
| ◌ Tracker operations | https://flow-next.dev/integrations/tracker-operations/ |
| ◌ Platform matrix | https://flow-next.dev/integrations/platforms/ |
| ◌ Overview | https://flow-next.dev/reference/ |
| ◌ All skills | https://flow-next.dev/skills/ |
| ◌ Flow | https://flow-next.dev/skills/flow/ |
| ◌ Strategy | https://flow-next.dev/skills/strategy/ |
| ◌ Prospect | https://flow-next.dev/skills/prospect/ |
| ◌ Chart | https://flow-next.dev/skills/chart/ |
| ◌ Capture | https://flow-next.dev/skills/capture/ |
| ◌ Refine | https://flow-next.dev/skills/refine/ |
| ◌ Plan | https://flow-next.dev/skills/plan/ |
| ◌ Work | https://flow-next.dev/skills/work/ |
| ◌ Make PR | https://flow-next.dev/skills/make-pr/ |
| ◌ Resolve PR | https://flow-next.dev/skills/resolve-pr/ |
| ◌ Plan Review | https://flow-next.dev/skills/plan-review/ |
| ◌ Impl Review | https://flow-next.dev/skills/impl-review/ |
| ◌ Spec Completion Review | https://flow-next.dev/skills/spec-completion-review/ |
| ◌ QA | https://flow-next.dev/skills/qa/ |
| ◌ Visual | https://flow-next.dev/skills/visual/ |
| ◌ Prose | https://flow-next.dev/skills/prose/ |
| ◌ Setup | https://flow-next.dev/skills/setup/ |
| ◌ Prime | https://flow-next.dev/skills/prime/ |
| ◌ Audit | https://flow-next.dev/skills/audit/ |
| ◌ Features | https://flow-next.dev/skills/features/ |
| ◌ Map | https://flow-next.dev/skills/map/ |
| ◌ Memory Migrate | https://flow-next.dev/skills/memory-migrate/ |
| ◌ Sync | https://flow-next.dev/skills/sync/ |
| ◌ Deps | https://flow-next.dev/skills/deps/ |
| ◌ Worktree Kit | https://flow-next.dev/skills/worktree-kit/ |
| ◌ Tasks & Inventory | https://flow-next.dev/skills/flow-next/ |
| ◌ Drive | https://flow-next.dev/skills/flow-next-drive/ |
| ◌ Overview | https://flow-next.dev/subagents/overview/ |
| ◌ Planning Scouts | https://flow-next.dev/subagents/planning-scouts/ |
| ◌ Readiness Scouts | https://flow-next.dev/subagents/readiness-scouts/ |
| ◌ Execution & Review | https://flow-next.dev/subagents/execution/ |
| ◌ flowctl CLI | https://flow-next.dev/flowctl/cli-reference/ |
| ◌ Configuration | https://flow-next.dev/flowctl/configuration/ |
| ◌ Review backends | https://flow-next.dev/reference/review-backends/ |
| ◌ Receipts | https://flow-next.dev/reference/receipts/ |
| ◌ Spec schema & ids | https://flow-next.dev/reference/spec-schema/ |
| ◌ Standing criteria | https://flow-next.dev/reference/standing-criteria/ |
| ◌ Glossary | https://flow-next.dev/reference/glossary/ |
| ◌ Troubleshooting | https://flow-next.dev/reference/troubleshooting/ |
| ◌ FAQ | https://flow-next.dev/reference/faq/ |
| ◌ Benchmarks | https://flow-next.dev/project/evidence/ |
| ◌ Changelog | https://flow-next.dev/releases/changelog/ |
| ◌ Compatibility & migration | https://flow-next.dev/releases/compatibility/ |
| GitHub ↗ | https://github.com/gmickel/flow-next |
| Discord ↗ | https://discord.gg/f3DYq8AAm5 |
| X ↗ | https://x.com/gmickel |
| Docs | https://flow-next.dev/introduction/ |
| Orchestration | https://flow-next.dev/guides/model-routing/ |
| Autonomy | https://flow-next.dev/autonomy/going-autonomous/ |
| Cookbook | https://flow-next.dev/guides/cookbook/ |
| Evidence | https://flow-next.dev/project/evidence/ |
| v9.17.0 changelog | https://flow-next.dev/releases/changelog/ |
| GitHub ↗ | https://github.com/gmickel/flow-next |
| proof | https://flow-next.dev/reference/receipts/ |
| Install Flow-Next → | https://flow-next.dev/install/ |
| Your first 30 minutes | https://flow-next.dev/first-30-minutes/ |
| See every number → | https://flow-next.dev/project/evidence/ |
| The plain-agent results → | https://flow-next.dev/project/evidence/#vs-plain-claude-code |
| The superpowers results → | https://flow-next.dev/project/evidence/#vs-superpowers |
| The pstack results → | https://flow-next.dev/project/evidence/#vs-pstack |
| How teams run it → | https://flow-next.dev/guides/for-teams/ |
| The field case → | https://flow-next.dev/autonomy/goal-mode/#field-case-67-prs-from-one-goal |
| The plain-agent results → | https://flow-next.dev/project/evidence/#vs-plain-claude-code |
| Walk it once in 30 minutes → | https://flow-next.dev/first-30-minutes/ |
| arXiv 2603.24755 ↗ | https://arxiv.org/html/2603.24755v1 |
| Everything reaches your queue checked as hard as its risk calls for. Review goes by risk. Changes that touch persisted or shared state, concurrency, security or data layout, and multi-file features, get three reviewers from another model family; a small change in one area gets one, and a small wording or display fix skips review and records why. Turn review off and the run says so in its report. ✓ Attended, you get the result first, then one review and one re-review of the fixes. Unattended, it keeps fixing and re-reviewing until SHIP. Every done carries commits and test evidence. review (by risk) 3 reviewers first pass NEEDS_WORK fix + re-review SHIP flowctl done evidence recorded | https://flow-next.dev/understand/how-work-gets-proven/ |
| Open a PR that already makes its argument. The reviewer's first screen is the reasoning behind the change and the criteria it claims to satisfy. ✓ The pull request arrives explaining itself: which acceptance criterion each change satisfies, what is still open for a human, what deliberately did not change. An unattended run adds every decision it made on your behalf, with the evidence. GitHub PR Why · Review focus · Needs you Shape · User impact · Reading order Tradeoffs + Decisions · Verified | https://flow-next.dev/skills/make-pr/ |
| Decide what to build before anyone builds it. An idea too big to write down gets charted one decision at a time; a conversation becomes a spec; a product owner and an engineer refine it in their own sessions on one file, when a decision is open. ✓ Chart resolves decisions against evidence, capture writes source-tagged criteria with a read-back, refine (interview) runs one question pass on the spec, focused by whoever runs it, or a read-only research pass, prospect ranks what to build next. /flow-next:chart resolve D-IDs /flow-next:capture spec + read-back /flow-next:refine any scope | research | https://flow-next.dev/skills/chart/ |
| Your team's context lives in the repo. The reasons behind the code sit next to the code. A new teammate, or the next agent run, starts from what the last one learned. ✓ A review correction becomes a lesson the next task can read. Specs, decisions, glossary, and memory stay in your repository. .flow/specs/ intent + criteria .flow/memory/ lessons that stuck GLOSSARY.md your domain words | https://flow-next.dev/understand/how-it-compounds/ |
| Prove it in the running app, not by reading the source. Live QA drives the app the way a user would, from the spec's own criteria, and files what it finds with screenshots and a verdict you can audit. ✓ The QA skill is forbidden from passing by reading code; a committed feature map records how each surface is reached so every run starts warm. /flow-next:qa fn-12 P0 0 · P1 1 · P2 2 6/6 R-IDs exercised · SHIP | https://flow-next.dev/skills/qa/ |
| Hand over as much as the receipts have earned. Start by watching a single task run. Move up a rung when the receipts have earned it, and step back down whenever you want. ✓ One dial from a supervised pair to a loop draining the backlog overnight. The same gates apply at every rung, and each records its verdict or its skip. Unattended, it never asks: it decides from evidence and writes each decision into the PR. /flow-next:flow fn-12 you watch /flow-next:flow --auto it runs unattended a bot or a cron it repeats | https://flow-next.dev/autonomy/going-autonomous/ |
| Choose the model for each job. Name a model per role once in your CLAUDE.md, or say it in the prompt for a single run. Pick another family for review and the model that wrote the diff never reviews it. ✓ A tier line in your instruction file, a --review= flag, or a sentence in the moment. Leave it unset and everything runs on the session model. spec -> the model you trust for judgment work -> a capable model for this spec review -> a different family | https://flow-next.dev/guides/model-routing/ |
| One way of working, from a solo Sunday to a rollout. The same rails carry a solo developer on a Sunday and a fifty-person organisation on a rollout. The spec is the handover object, and it reads the same to product, engineering, and the next agent run. ✓ Product and engineering refine the same file, each with their own focus, tracker projection to Linear, GitHub, GitLab or Jira, and a one-hour discovery interview that yields eight to eleven implementation-ready specs. product -> Goal, Boundaries engineer -> Architecture, Edge cases both -> Acceptance Criteria | https://flow-next.dev/guides/for-teams/ |
| Your process outlives your agent. Switch harness, model, or vendor and nothing has to move. The specs, the gates, and the record of what happened are files in your repository. In a harness that can dispatch subagents, the same routing runs across models in-host with no bridge at all. ✓ The same specs, gates, receipts, and task state across harnesses. Everything in your repository, and nothing outside it. Claude Code -> write the spec Codex -> implement Cursor -> review same .flow/ state, every role uninstall: rm -rf .flow/ | https://flow-next.dev/understand/architecture/ |
| Hand your validation team the record it needs. For regulated teams (GxP, medical devices, finance): opt in, and Flow-Next commits every review round, QA capture and test record per spec, holds implementation until a human approves the spec's pull request, and prints a traceability matrix from requirement to test, QA, review and approval. ✓ Three explicit settings, all off by default. It feeds your team's validation and leaves the release decision with your QS; it makes nothing compliant on its own. evidence.persist on committed evidence per spec spec.approval pr no build before an approved spec PR flowctl trace fn-12 R-ID -> tests, QA, findings, approvals | https://flow-next.dev/guides/regulated-environments/ |
| How composing works, with recipes you can use → | https://flow-next.dev/understand/compose-the-pipeline/ |
| Flow --auto, the build loop One invocation carries one blessed spec hop after hop to a pull request, ready unless something is left for a human, and ends with a verdict your host can act on; --tick runs one hop for hosts that loop. /flow-next:flow --auto Classifies each hop from the same routing references the attended conductor reads: plan, plan review, work, QA, make-pr. Decides per spec whether to plan first, and prints why. Never asks: it decides from evidence, keeps fixing until review says SHIP, and writes every decision into the PR. With --until=merge it makes reversible calls inside the spec itself, stops before the merge on any call that is yours, and merges after a 10-minute window for review bots. Backlog mode triages the whole board in dependency order, one item per run. | https://flow-next.dev/autonomy/pilot/ |
| Land, the ship loop Babysits the one pull request you name until it merges, on the repo's own rules and with your authorization. /loop 15m /flow-next:land 123 Makes one CI fix or one flake rerun per failure, then stops and says so. Resolves review threads through resolver agents, replies with the fix and the line. Squash-merges with a head-commit match only when you authorized it, and writes nothing to the repository afterwards. | https://flow-next.dev/autonomy/land/ |
| Forge ↗ | https://www.killenberger.com/#forge |
| Daniel Killenberger ↗ | https://www.killenberger.com/ |
| @patrickmichalina · issue #5 ↗ | https://github.com/gmickel/flow-next/issues/5#issuecomment-3734228766 |
| @possibilities · PR #95 ↗ | https://github.com/gmickel/flow-next/pull/95 |
| @raydocs · issue #4 ↗ | https://github.com/gmickel/flow-next/issues/4 |
| Evidence and evals → | https://flow-next.dev/project/evidence/ |
| Flow Hand it anything. It picks the route from the cells below, runs it, and stops at your decision. The one command a first run needs. /flow-next:flow | https://flow-next.dev/skills/flow/ |
| Chart Resolve one decision at a time until an idea is briefable. /flow-next:chart | https://flow-next.dev/skills/chart/ |
| Capture Conversation to spec, source-tagged, with a read-back. /flow-next:capture | https://flow-next.dev/skills/capture/ |
| Refine (interview) One question pass on the spec, optionally focused by a scope, or a read-only research pass. /flow-next:refine | https://flow-next.dev/skills/refine/ |
| Prospect & Strategy Ranked candidates grounded in the repo; the strategy anchor they answer to. /flow-next:prospect | https://flow-next.dev/skills/prospect/ |
| Plan Optional task graphs when you ask for a plan or a task breakdown. /flow-next:plan | https://flow-next.dev/skills/plan/ |
| Work One task inline, several in parallel lanes, reviewed by risk. /flow-next:work | https://flow-next.dev/skills/work/ |
| Reviews Plan, implementation, completion; cross-family, receipted. /flow-next:impl-review | https://flow-next.dev/skills/impl-review/ |
| Live QA & Drive Run the app like a user; file what breaks with evidence. Off, on, or auto. /flow-next:qa | https://flow-next.dev/skills/qa/ |
| Make PR A PR that argues its case, with the review evidence attached. /flow-next:make-pr | https://flow-next.dev/skills/make-pr/ |
| Resolve PR Review threads triaged, fixed or answered, and resolved by agents. /flow-next:resolve-pr | https://flow-next.dev/skills/resolve-pr/ |
| Flow --auto & Land The build loop that carries one ready spec to a PR without asking, and the ship loop that merges it. /flow-next:flow --auto | https://flow-next.dev/autonomy/pilot/ |
| Model routing Four tiers, a routing block, bridges to other CLIs. reviewer: | https://flow-next.dev/guides/model-routing/ |
| Tracker sync Project specs to Linear, GitHub, GitLab or Jira. Projection, never coordination. /flow-next:tracker-sync | https://flow-next.dev/integrations/tracker-sync/ |
| Memory & Audit Lessons that stuck, audited against the code that moved. /flow-next:audit | https://flow-next.dev/skills/audit/ |
| Prime & Features Agent-readiness assessment; a user-POV feature map QA reuses. /flow-next:prime | https://flow-next.dev/skills/prime/ |
| Visual & Prose See any spec, plan or diff visually; one prose contract for every artifact. /flow-next:visual | https://flow-next.dev/skills/visual/ |
| Map A semantic feature index scouts and prime anchor to. /flow-next:map | https://flow-next.dev/skills/map/ |
| Plan-sync Downstream task specs updated when the code moved under them. /flow-next:sync | https://flow-next.dev/skills/sync/ |
| Deps & Worktrees Execution order across specs; isolated worktrees for parallel work. flow-next-deps | https://flow-next.dev/skills/deps/ |
| Setup & Uninstall A copy-less install; removal is one directory. /flow-next:setup | https://flow-next.dev/skills/setup/ |
| All skills, with what each prints → | https://flow-next.dev/skills/ |
| Set up a review command → | https://flow-next.dev/reference/review-backends/#review-command |
| What changed, and how to upgrade | https://flow-next.dev/releases/changelog/#flow-next-9-17-0 |
| 9.16.0 Reviews on the newest models A reviewer set up without naming a model runs the newest reviewer models at the best review per unit of cost. The most expensive tiers run only when you pin them. | https://flow-next.dev/releases/changelog/#flow-next-9-16-0 |
| 9.15.0 Idea intake is safe to rerun A rerun after a failed delivery link finds the parent epic the first run created, by its label, instead of making a second one. | https://flow-next.dev/releases/changelog/#flow-next-9-15-0 |
| 9.14.0 Tracker rejections say why When Jira or GitLab turns a request down, the error adds the tracker's own reason after the status. A token command can stay in one machine's environment, out of the committed config. | https://flow-next.dev/releases/changelog/#flow-next-9-14-0 |
| All release notes → | https://flow-next.dev/releases/changelog/ |
| Your first 30 minutes → | https://flow-next.dev/first-30-minutes/ |
| How teams run it → | https://flow-next.dev/guides/for-teams/ |
| Org-wide deployment → | https://flow-next.dev/install/ |
| Install → | https://flow-next.dev/install/ |
| Read the docs | https://flow-next.dev/introduction/ |
Viewport: width=device-width, initial-scale=1.0
Robots: index,follow