cd /news/developer-tools/ducklab-dev-harness-that-built-itsel… · home topics developer-tools article
[ARTICLE · art-107874] src=github.com ↗ pub= topic=developer-tools verified=true sentiment=↑ positive

Ducklab – dev harness that built itself: 416 runs, $176, local models

Ducklab, an Apache-2.0 self-hosted software development harness built with a Go engine, CLI, and desktop app, has completed 416 runs at a total cost of $176 using local models first, and has developed itself through 111 accepted tasks. The harness, designed for multi-LLM use, ensures verdicts are based on exit codes rather than model opinions, and supports local models via llama.cpp and vLLM alongside OpenAI-compatible and Anthropic endpoints. It operates over MCP and records all decisions, with a gate that reproduces tests from clean checkouts before any commit is accepted.

read12 min views1 publishedAug 23, 2026
Ducklab – dev harness that built itself: 416 runs, $176, local models
Image: Michielbdejong (auto-discovered)

A full-cycle software development harness that is multi-LLM by default and honest by construction.

In one block: self-hosted development harness (Go engine + CLI + desktop, Linux first) · brief → requirements → spec → plan → build → review → release · verdicts are exit codes, never model opinions · local models first (llama.cpp, vLLM) beside any OpenAI-compatible or Anthropic endpoint · operable by humans or by other agents over MCP with recorded, attributed decisions · Apache-2.0 · develops itself (the run records in .ducklab/

are the receipts). Agents: start at AGENTS.md and

.

llms.txt

You give it a brief. It writes requirements, a spec and a plan; builds tasks with one model or several arguing; runs your project's real test gate; and stops for you before anything is committed. Every model call is logged. No model ever decides a verdict.

A real council intake, recorded live and sped up: the architect streams the draft, a different model reviews it, the budget ticks in cents — and the run stops at your gate. Total cost of what you just watched: $0.07.

It was built for local models first — the two that built most of it are a vLLM box on the LAN and a llama.cpp server on localhost, both priced at zero — and hosted models sit beside them in the same roster, measured by the same evidence.

Most agentic coding tools assume one strong model and trust it. Ducklab assumes several cheap models and trusts none of them:

The gate decides, never a model. A verdict is a command's exit code. A test-first run measures a greenbaseline before any test is written, the redover the new test after, and every acceptreproduces the gate from a clean checkout of the committed sha— nothing lands that did not reproduce, and an accept whose reproduction fails takes its own commit back.** Decorrelation everywhere.A different model reviews; a reviewer never learns who wrote the code (absent from the payload, not hidden in the UI); tournament judges choose blind; council critics read the draft, not each other.Work is a contract. A task's deliverables are the implementer's numbered checklist; it reports on each by number, the reviewer checks each against the diff, and an undelivered item summons therubber duck**— an advisor seat that wakes only on measured distress (brake refusals, failure streaks, red gates) and answersnone

, a note that sends the implementer straight back to work, orstop

.Seats are chosen on evidence. Every duckling carries a scorecard — in-seat pass rate from your own runs, cost per run, coding index — and the roster board suggests seats from it, with the ranking criteria yours to reorder. Suggestions are rare and justified: pass rates rank by their Wilson lower bound, three runs minimum, locals never win on a $0 price.Nothing is unbounded. Turns, tokens, cost, wallclock, tool output, shell commands — every ceiling visible and liftable mid-run, on the record.Your documentation is not bounded by the model's window. Attach a wiki to a stage and a big seat reads it whole; a small seat gets each document digested to fit, the full text oneref_read

call away, and the gate names any document nobody opened. A 32k local model can be briefed by a quarter-million characters of reference material — the harness carries the working memory.

The record does not round up: every run with its verdict, its cost, and whether its accept reproduced green from a clean checkout.

And the existence proof: ducklab is developed inside ducklab. The plan, the bugs, the releases and 111 accepted tasks went through its own loop, driven by the same local and hosted models it measures — most recent features (the escalation suggestions, the acceptance receipts, the MCPB release packaging, the multimodal chat) were built by the duck, gated by a person. Don't take the claim on faith:

git clone https://github.com/jrullan/ducklab && cd ducklab
go build -o ducklab-cli ./cmd/ducklab
for r in .ducklab/runs/*/receipt.json; do ./ducklab-cli proof verify "$r"; done

Receipts ship with every accept since v0.7.0: the committed sha, the gate command, its exit code, and the clean-checkout reproduction verdict — facts a third party re-derives, never assessments.

v0.7.0, moving fast — seven releases in the first three weeks. Seven stages, five modes, the roster board with evidence and suggestions, reference documents with automatic digestion, skills managed from the desktop, a seated consultant you can chat with (images included, vision verified before they are sent), bugs with screenshot evidence, adopt surveys with a deterministic coverage check at the gate, provider-aware queueing that says why a run waits, escalation suggestions when a seat measurably hits its ceiling, exportable acceptance receipts with ducklab proof verify

, releases, autopilot, a CLI, a desktop app, and an MCP server — in the official MCP registry as io.github.jrullan/ducklab

— that lets another model operate the whole loop with recorded, attributed decisions.

docs/status.md tracks all acceptance criteria and does not round up. Where code and spec differ, the difference is recorded in

.

docs/decisions/

Needs Go 1.25+, Node 22+ for the desktop, and git.

The CLI and engine are pure Go. The desktop is a Wails v3 app and needs the GTK/WebKit development packages:

sudo apt install libgtk-3-dev libwebkit2gtk-4.1-dev   # Debian/Ubuntu names
make desktop && make install

On Ubuntu 24.04+ the desktop also needs an AppArmor profile — see decision 0003 and packaging/apparmor/

.

xcode-select --install    # the desktop build links against WebKit
brew install go node
make desktop && make install

Honesty note: ducklab is developed and exercised daily on Linux. The CLI and engine compile-check for darwin/arm64

on every make cross

, but no desktop build has been verified on a Mac yet — the first person to try it is the test, and make install

gives you the CLI and engine either way. Please report whatever breaks.

make install

installs to ~/.local/bin

— make sure it is on your PATH

. It warns when the desktop binary predates frontend/src

, because it will happily install a stale one.

To exercise the frontend against the lightweight fake engine, run the engine and Vite in separate terminals, then open the browser with its connection details:

go run ./cmd/fake-engine --port 8787 --token fake-token
npm run dev --prefix frontend

The engine

and token

query parameters are available only in Vite dev builds. They can also be supplied as VITE_DUCKLAB_ENGINE

and VITE_DUCKLAB_TOKEN

environment variables. The desktop shell continues to use its injected window.ducklab

connection.

What it is
ducklab-engine
The daemon. Owns every run. Binds 127.0.0.1 only, bearer token rotated each start.
ducklab
The CLI client. Holds no state; it asks the engine.
ducklab-desktop
The desktop app. Also a client, also holds no state. Starts (or adopts) the engine itself.

Provider keys come from the engine's environment at call time — export them before it starts, or launch the desktop through a wrapper that loads them from your keyring. The app tells you when the engine it adopted is missing a key this app has, with the restart button beside the words.

From the desktop: Projects → New project, then Cycle → Draft it. From a terminal:

cd ~/dev/myproject
git init                                    # ducklab needs a git repo
ducklab project init --name MyProject       # auto-starts the engine if none is running

ducklab intake --from brief.txt             # brief        → requirements
ducklab spec                                # requirements → spec
ducklab plan                                # spec         → milestones and tasks

ducklab run T-001                           # build it
ducklab run accept r-20260729-...           # commit it

ducklab review T-001                        # read the commit
ducklab release plan --bump minor           # what shipped

Each stage writes a .proposed

file first and waits for you. accept

promotes it; reject

restores exactly what the run wrote and nothing else; "request changes" sends any draft — spec, plan, release notes — back with your note. Nothing is committed without you (or without the autonomy level you explicitly granted).

Reference documents ride any stage: --ref ~/wiki/product/

(or the attach door in the desktop) loads files or whole directories as background for the architect — grounded by two rules the prompt states outright: the approved requirements own the scope, and where a reference and the code disagree, the code is the truth. When the corpus outgrows the seat's context, each document is digested once (cached by content hash), the full text stays reachable through the ref_read

tool, and the proposal card lists any document no seat ever opened.

Adopting an existing codebase works the same way: intake reads the code and writes as-built requirements, the spec marks its sections as-built

, and the plan stays deliberately empty — new work then enters through bug reports and plan amendments, which is how ducklab itself is developed.

Your project declares its own truth in .ducklab/project.toml

: the gate ([verify]

— with link_deps

and setup

for what a clean checkout needs), how the app launches ([run]

with a preflight), and how the project's own binaries are rebuilt ([install]

) so the whole loop runs without leaving ducklab.

Gate and shell process trees always receive DUCKLAB_RUN_ID

and DUCKLAB_PROJECT_ID

. For example, excercise-tracker can use DATABASE_URL=test_db_${DUCKLAB_RUN_ID}

in [verify].tests

, and a compose preflight can use ${DUCKLAB_PROJECT_ID}

as its per-run project name. Ducklab guarantees identity only; provisioning and teardown remain the project's.

ducklab provider set openrouter --url https://openrouter.ai/api/v1 \
                                --key-env OPENROUTER_API_KEY
ducklab duckling set pato-sonnet --provider openrouter \
                                 --model anthropic/claude-sonnet-4.5 \
                                 --roles reviewer,judge --context 200000 \
                                 --cost-in 3.0 --cost-out 15.0
ducklab duckling test pato-sonnet --prompt "say OK"

--key-env

is the name of an environment variable, never a key. No key is written to config, sent over the API, or kept in shell history.

Seats are argued with evidence: pass rates from your own runs, cost per run, coding index — suggestions justified, never imposed.

The desktop's Roster view is where seats are assigned: drag from the Flock onto a mode's seat, globally or per project, with each duckling's evidence on the card and the engine's suggestions beside the seats. Coding / intelligence / agentic indices come from OpenRouter's benchmarks endpoint when a duckling lives there; your own runs supply the rest.

The same machinery on real work: a council revising ducklab's own spec, 4.5M tokens in, d once on a budget it asked to lift.

ducklab run T-001 --mode <mode>

Mode What it does
solo
One duckling. The yardstick everything else is measured against.
pair
Implementer and reviewer, decorrelated. Between them the advisor — the rubber duck.
tournament
Contestants build the same task in isolated worktrees; a judge picks, blind.
split
An architect decomposes; subtasks run in parallel; integration is file copies, no model involved.
council
Several models on one document, for intake, spec, plan and review. One drafts, the others critique blind, the first revises.

These are load-bearing, not preferences.

A model never decides a verdict. A gate is a command's exit code.A green candidate is applied byte-for-byte. Nothing is re-generated after it passed.**A reviewer never learns who wrote the code.**Nothing lands that did not reproduce from a clean checkout.**A reject undoes what the run wrote, and nobody else's work.****Nothing is unbounded.****Secrets never touch project state.**The engine is loopback-only. There is no remote mode.

A skill is a directory with a SKILL.md

— under .ducklab/skills/

for one project, or in the machine-wide skills directory to serve every project (project shadows global on a name collision). The documentation-only form has no script and is the default: a recipe a model reads and follows. The architect reads survey guides before an adopt (skill_list

is in its prompt), the consultant reads them in chat, and only the implementer can skill_run

an executable one.

Skills are administered from the desktop (gear → Skills): list with scope badges and validation problems, read, edit the whole SKILL.md

, run with arguments, delete. A skill a duckling writes during a run shows there greyed pending acceptance

until its run is accepted — proposing a skill goes through the same gate as proposing code.

ducklab skill new house-style
ducklab skill run changelog-entry --arg summary="..."

Every project seats a consultant (a Common seat on the roster board): the model behind the "chat about this" doors and the free-form chat in the guide rail. It reads the code, the runs, the boards and the skills — never writes — and takes images: paste a screenshot of a broken view and ask. Vision is verified, not assumed: a declared-vision seat is probed with a real image request once, and a text-only seat refuses the paste with words instead of hallucinating an answer.

Asked to sell the product, the seated consultant read the repo and wrote this pitch itself. We kept it.

ducklab mcp serve

exposes the whole loop over stdio as an MCP server: an external model reads each result, decides gates (with a required, recorded reason — decisions land as approved_by: mcp:<client>

, never as "human"), answers questions, files bugs, amends plans and starts work. The engine's next

lists are the law: an operator cannot take an action a person could not.

See CONTRIBUTING.md — how to build, how the tests guard the architecture, how work flows through ducklab's own loop, and where to start. The short version:

make            # vet, test, build the frontend
go test ./...   # 38 packages
cd frontend && npx vitest run

License: Apache-2.0. Contributions are accepted under the same terms (§5 of the license — no CLA). The Ducklab name and the duck are the maintainer's (§6).

The code implements a written specification, in this repo: docs/spec/ (00-VISION through 08-DESKTOP-UI) is the

normative layer — vision, invariants, protocol contracts, acceptance criteria. What the system IS today lives in

.ducklab/docs/

— the as-built requirements, spec and plan the loop itself maintains, each version signed at a human gate. Where the two differ deliberately, the difference is recorded in ; the diff between them is the roadmap, and the alignment stage computes it.

docs/decisions/

── more in #developer-tools 4 stories · sorted by recency
── more on @ducklab 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
Live at https://your-agent.zahid.host
Get free account → Pricing
from €0/mo · no card required
LIVE [news/ducklab-dev-harness-…] indexed:0 read:12min 2026-08-23 ·