Building in public · roadmap, code and failures

Describe the work once.
Then close the laptop.

The weekly research, the recurring digest, the sweep nobody gets to, the ticket that has waited a quarter. Describe it once in plain English and a team of agents picks it up — product, architect, developer, QA — with an independent manager checking each step before the next one starts. The whole record lands in your git repository.

Open source Self-hostable SaaS & on-prem
forge · research run
$ forge run "size the RU market for BPA tooling"
// measured, not modelled

Twelve weeks of Forge building Forge.

One workflow — the one that builds this platform — counted from the control plane's own records.

0
Handoffs judged by a management agent
0
Stopped before the next role saw them
0
Pull requests merged by agent teams
0 min
Median run, dispatch to finished

Every run the workflow made in that window — not a sample, not a pilot, not a case study we cannot show you. A person performed every merge. OpenSSF Scorecard 10/10, MIT licensed.

// why forge

What nothing else on your shortlist does.

Five things that hold against every tool we get compared with.

independent review

A second agent judges every handoff

Everywhere else, the agent that asked for the work also decides whether it was any good. Forge puts an independent management agent on each edge between roles: it can pass the work, send it back with specific rework instructions, or escalate to a person — before the next role builds on it.

pass
return for rework
escalate
Learn more →
provenance

The reasoning is committed, not summarised

Every run leaves a branch in your repository: the spec, the brief and each role's handoff as commits, authored by the persona. Readable with git log, months later, by someone who was not there.

git log Co-authored-by
Learn more →
no graph to draw

You describe the team, not the wiring

No canvas, no nodes, no edges to keep in sync. Name the roles, say what done looks like, and the engine assembles the run — then keeps it working when the job changes.

Learn more →
your keys, your metal

Your hardware, your provider pool

Self-hosted runners keep execution on your machines. Inference routes across a provider pool you control — priority weights, automatic failover, any OpenAI-compatible endpoint, including models you host yourself. Agents never hold your keys.

Anthropic OpenAI Google Ollama +3
Learn more →
not a coding tool

The same engine runs work that is not software

Research that verifies its own sources, a recurring digest, a content desk, an ops rotation. Same roles, same review loop, same audit trail — software delivery is one archetype, not the product.

Learn more →
// head to head

How we read against the tools you are also looking at.

Each page quotes the other product in its own words, then names the thing it does not do.

All comparisons →

// how it works

Describe the work.
Get output that was checked.

A team of agents runs the job. A supervisor reads every handoff and decides whether the next step gets it — so weak work stops instead of travelling.

Your brief forge-pipeline
01
Scope Requirements
02
Plan Design & API
03
Do the work Code & tests
04
Check it Tests & review
You approve Ready
// get started

Ready to put your team on auto?

Describe your workflow in plain English. Forge runs it — across SDLC, research, content, and ops.

We use this address only to contact you about Forge — no list, no sharing. Ask us at [email protected] and we delete it.