Anvil CoderLog in

Forge something mighty.

The problem with coding agents is not their code. It is that you cannot merge it.

Anvil Coder takes a ticket from your backlog and delivers a pull request in which every step has been checked and can be traced before you see it. The backlog shrinks. You only decide.

Animation: a ticket is cut into four work steps; each step passes the Anvil Coder gate as one commit, a red gate sends one step back once, the commits land in a draft pull request that waits for your review, and eight phases close one after another.Ticket #482 · CSV export of ordersJira · GitHub · GitLab · YouTrack · internal boardCut into work stepsStep 1 · endpointStep 2 · serviceStep 3 · testsStep 4 · docsGate red: step runs againGuardrails · gate: build + testDraft pull requestfeat: export endpointfeat: csv servicetest: export casesdocs: export guideCommit and trail per stepReady for your reviewEight phases, each behind a barrierREQARCHIMPLRECONTESTVERIFYDOCSCONS

Sound familiar?

Anvil Coder turns that around: a ticket is split into small work steps, each runs under the same guardrails, leaves its own commit and has to pass a gate before the next phase begins. What reaches you is a draft pull request, and you read on the commit which assignment it came from.

Watch the work happen, decide at the end

The cockpit shows every work step as it completes, runs or waits, with its cost next to it. The pull request lands in your code host, where you review anyway.

Anvil Coder run overview showing completed, running and pending work steps (German interface, example data)
The run overview in the cockpit, example data, German interface.

A real run, measured

A Laravel bug from the tracker became three commits in 24 minutes: the fix, the follow-up in six dependent files and a regression test, with no human intervention. The sandbox lacked the PHP toolchain, so the run reported the test as unverified instead of claiming green.

Aug 31, 2026 (one run, ~24 minutes)

Read the run step by step

Which tools can you connect?

Your model provider, your code host, your tracker: Anvil Coder is built to fit into the landscape you already have.

Model connections

  • Anthropic
  • OpenAI
  • Groq
  • Gemini
  • Mistral
  • Ollama

Built with

  • Java
  • Spring Boot
  • Angular
  • Astro
  • PostgreSQL
  • NATS
  • Kubernetes
  • Flux CD

Integrations and operations

  • GitHub
  • YouTrack
  • Keycloak
  • nginx
  • Prometheus
  • Grafana
Explore integrations and requirements

You keep your process: your repository on GitHub, GitLab or Bitbucket, your tickets in the internal board, Jira, YouTrack, GitHub or GitLab, your model provider with your own key. Java with Spring Boot is the most thoroughly proven path;here is how you bring your stack.

How it worksOne ticket from the plan through the commits and gates to the pull request.GuardrailsYour architecture rules as a condition on every step, not a hint in the wiki.SecurityHosted in Switzerland, your organisation in every record, your token stays with the coordinator.

What does Anvil Coder cost?

One Team subscription per organisation covers ten seats and includes runtime credit for your runs; beyond that you pay for what you use. Every project shows its own costs, and a cap per run holds before it hurts.

See plans and usage costs

What people ask before their first run

Does Anvil Coder merge automatically?
No, the result arrives as a draft pull request in your repository, and a human merges it. How much you want to approve during the run is a per-project setting.Details: Review and supervision
What if the result is no good?
During the run, the consistency pass sends affected steps into capped repair rounds; afterwards, a comment on the pull request becomes a follow-up run. You write what is missing instead of rebuilding it yourself.The flow as a whole: How it works: phases and slices · Integrations: the feedback channel on the pull request
Does it run in Switzerland?
Yes, the service runs on the Infomaniak Public Cloud in Switzerland. Model requests go to the provider you configure, and the result goes wherever your repository lives.Where the data lives: Security: where the data lives
How do I keep control?
Supervision is a per-project setting: from “every action needs an approval” to “only the consequential ones”, or no intermediate approval at all. The merge stays with you either way.Why that is the real question: For product teams: hand over the execution, keep the decision

All questions and answers

The method behind the workflow: Spec-Driven Development · GitHub Spec Kit.