The AI coding agent for product engineers · Excalibur

GitHub Get started

OSSExcalibur Core is open-source · Apache-2.0<br>The AI coding agent for product engineers.

Most agents write code. Excalibur knows the whole product cycle — discover, build, verify, ship.

$npm install -g @excalibur-oss/excaliburApache-2.0

Get startedI'm a manager

excalibur — m-shell<br>›Add rate limiting to the public API<br>L4 — Full agentic · kimi-k2.7-code<br>✓Plan4 steps

⠋Implementapi/limiter.ts<br>│└ edit api/limiter.ts+38 −4<br>│└ ▸ pnpm test apirunning…

○Verify

○Document

○Review

○Pull Request

████░░░░1m12s·$0.42·standard-safe·no push

With Excalibur you can:<br>Time-travel your work·Discover what to build·Run it from a local board

01What you can doExcalibur enables you to build like never before.<br>Five capabilities built right into the engine — the moats that hold up against any other AI coding tool.

the immutable event logTime-travel your work

Every run is an immutable, append-only event log — so you scrub it like a video and branch a new run from any step. The stream IS the run: it renders the live view, the replay, the dashboard and the audit trail, byte-identical. Nobody else is built this way.<br>Fork from cache: the good prefix replays for free; only what changed re-runs<br>Live rewind mid-session (Esc-Esc)<br>Live = replay = dashboard = audit — one source of truth

replay · run_4f2a<br>step 06 / 14 · Implement+38 −4<br>fork from here ⑂

planimplementtests failreview<br>live = replay = dashboard = audit · one immutable stream

the gate no one else hasDiscover what to build

Before a line of code, Discovery weighs scope, evidence and risk — and recommends build, validate, or don’t-build.<br>Product judgment, not just code generation<br>Turns an idea into a scoped work item<br>Kills bad work before it costs you

discovery · add AI invoice summaries<br>clarity<br>evidence<br>scope<br>risk<br>readiness<br>recommendation: needs_validation<br>→ thin evidence — validate with 3 design partners first · work item WI-142

local · no SaaSYour repo on a board

“excalibur serve” opens a task-first web dashboard — embedded in the CLI, one local process, no account. It’s that same event log, rendered for the browser. Terminal-only tools don’t have this.<br>Kanban work items → a run’s live checklist, patches, PRs and plans<br>Cost insights + a live swarm chronogram you can pause and replay<br>--write drives runs from the browser; --share mints a read-only link

dashboard · localhost:7878<br>backlog<br>WI-140 · webhook retries

running<br>WI-142 · invoice AI ◐

done<br>WI-138 · auth ✓

$2.40 today · 3 runs · live chronogram ▸▸▹ · share ↗ read-only

no lock-inAny model, any provider

Excalibur isn’t wired to one vendor — OpenAI-compatible (incl. Azure), Anthropic, Ollama and more. Pick a good + fast pair with a single key and switch anytime.<br>A frontier model for the hard parts, a cheap one for ghost-text — automatically<br>Bring your own key — it lives in an env file, never the repo<br>Switch model mid-session with /models

/models<br>●claude-opus-4good · default<br>○gpt-4o<br>○llama-3.3-70bfast · paired<br>○ollama/qwenlocal<br>good + fast from one key · no vendor lock-in

not just the diffThe whole product cycle

Most agents write code. Excalibur owns the bookends too — deciding what to build up front, then proving and auditing what shipped at the end.<br>A big task auto-sizes into a swarm of agents in isolated worktrees<br>An adversarial verification mesh + typed claims gate the finish<br>Tests and serious docs are phases, not afterthoughts

workflow · standard-feature<br>Discover→Plan→Build→Test ⛬→Docs ⛬→Review→Ship→Audit<br>→ sized to 3 agents — independent modules, isolated worktrees<br>✓ verification mesh · tests_passed · type_safe · no_secrets

All five — free, open-source, one npm install away.

For developers<br>02The whole cycleNot just code. The whole product cycle.<br>Chatbots autocomplete. Coding agents build and ship. Excalibur runs the entire cycle — deciding what to build up front, and auditing what shipped at the end — and it’s fully open source.

DiscoverPlanBuildTestDocumentReviewShipAudit<br>Excalibur<br>Coding agents

Coding agents take you from plan to ship. Excalibur owns the bookends too — deciding what’s worth building, and auditing what shipped.

Decide before you build<br>Discovery clarifies scope first — and can recommend not building it. Most tools start at the code.

Quality gates, built in<br>Every run plans, implements, tests and documents the change — tests and serious docs are phases, not afterthoughts.

Ship, review & audit<br>Pull requests, approvals and a full audit trail — not just a diff in your editor.

Every step configurableEvery action traceableEvery risky op approved

For developers<br>03QuickstartValue in minutes, not migrations<br>Install, run it in any repo, and start building — the first run sets itself up; then an agent plans, implements, tests and documents real changes in an isolated branch.

01Install · one binary, local-first

02Run excalibur · first run detects your stack & connects a model

03Describe a task · an agent...