Not just chat. Finished work.

Give OwlCoda the goal.Get back accepted work.

OwlCoda enters the real work environment, investigates what is actually happening, keeps ownership of the next step, verifies the result, and continues until the outcome is accepted. Delivery control is the core product; the CLI is one execution surface.

Start with delivery controlView the engine →
  • Project-owned truth
  • Evidence-linked
  • Human acceptance
  • Continuous delivery
owlcoda · project deliveryworkcase / equipment-release
$ npx owlrunkit project status --workspace .
# the project survives the current agent and session
Goalrelease equipment packageaccepted
Gapresolve BOM quantity driftowner assigned
Evidence14 source refs boundverified
Outcomecandidate ready for reviewhuman required
next → review the quantity decision and continue delivery.
owlcoda · delivery control

PROJECT TRUTH · DELIVERY LIFECYCLE

  • Goal contractone outcome · one ownerready
  • Delivery stateevidence linkedready
  • Business actionseparate authoritygate

In five seconds

From a goal to a verified result.

  • Starts witha project goal, not a temporary chat prompt.
  • Keeps movingthrough assignments, handoffs, rework, recovery, and one visible next step.
  • Proves donewith candidate identity, verification evidence, delivery state, and human acceptance.
  • Executes throughOwlCoda CLI or another bounded executor without surrendering project truth.

Primary product surface

Start with the project, not the prompt.

OwlCoda delivery control is powered by the public OwlRunKit engine. It preserves the goal, owner, candidate, evidence, handoff, recovery, and delivery state across agents and sessions.

Add delivery control to a project

npm install --save-exact [email protected]
  • Node ≥ 20.19
  • project-owned state
  • GPL source
  • no hidden dispatch
Step by step
  1. 01

    Install the exact project dependency

    npm install --save-exact [email protected]
  2. 02

    Preview safe onboarding

    npx --no-install owlrunkit bootstrap --workspace . --exact [email protected] --dry-run
  3. 03

    Apply when the project is idle

    npx --no-install owlrunkit bootstrap --workspace . --exact [email protected] --apply
  4. 04

    Read the durable project state

    npx --no-install owlrunkit project status --workspace .

OwlCoda CLI · execution surface

When the project needs action, the CLI goes to work.

The CLI connects models and tools to real execution. It is a strong expert surface, but it no longer carries the whole product story: delivery control remains above the session.

  • 42+ tools Bash, Read, Write, Edit, Glob, Grep, Task, MCP-served tools, agent dispatch, scheduling, plugins.
  • 69+ slash commands Inspect models, costs, budgets, context, sessions, skills, training — all from the same prompt.
  • Selection-first transcript Drag-select and copy work the way they do in any other terminal app.
  • Persistent sessions Every conversation lands under ~/.owlcoda/sessions/ and resumes with --resume <id>.
  • Sustained-runtime budget gate Long runs are checked against a context budget so a finished report can’t be silently truncated.

Slash commands · selected from 69+

  • /modelList configured models, switch active route.
  • /budgetInspect context budget and per-run caps.
  • /contextPrint current context window state.
  • /sessionsBrowse and resume saved sessions.
  • /resumeResume a previous session by id.
  • /trainingScore and export sessions for fine-tuning.
  • /traceOpen the trace inspector for the last run.
  • /perfShow per-tool latency and token spend.
  • /doctorProbe routes, keys, and runtime health.
  • /toolsList available tools, scoped or global.
  • /skillsManage learned skills and their triggers.
  • /tokensPer-message token accounting.
  • /costSpend by provider for the current session.
  • /configShow or edit ~/.owlcoda/config.json.
  • /agentsList available subagents and dispatch one.
  • /planEnter plan mode for multi-step work.
  • /clearClear the visible transcript (keep session).
  • /undoRoll back the last tool-applied change.
  • /openOpen admin UI route from the REPL.
owlcoda · session 7e3a~/work/owlcoda
$ /model
default → qwen3-coder (ollama · local)
heavy → kimi-for-coding (kimi · provider-native)
long-context → minimax-m27 (minimax · messages)
cheap → openrouter/auto (openrouter)
$ /model heavy
✓ active route → kimi-for-coding · ctx 128k · key from env.
$ patch the completion-guard false positive
Readsrc/native/conversation.ts412 L · 13.2 KB
Grep"completion-guard" → 6 matches3 files
Globsrc/native/**/*.ts47 paths
Editsrc/native/completion-guard.ts3 hunks proposed
Bashnpm test --silent14 passed · 2.1 s
Taskspawn review-agentin-progress
✓ patch applied · vitest 14/14 · session saved.

Bring your own backend

OwlCoda doesn’t ship a model. You point it at one.

Wire up local runtimes, your cloud API keys, or both. Endpoint and provider probes treat them with one mental model.

BackendKindEndpointWire formatStatus
OllamaLocalhttp://127.0.0.1:11434/v1OpenAI-compatible · everyday local routeauto-detect
LM StudioLocalhttp://127.0.0.1:1234/v1OpenAI-compatible · GUI runtimeauto-detect
vLLMLocalhttp://127.0.0.1:8000/v1OpenAI-compatible · GPU runtimeauto-detect
Custom localLocaluser-suppliedOpenAI-compatible · any /v1 endpointauto-detect
Kimi CodingCloudhttps://api.kimi.com/coding/v1provider-native · kimi-for-codingtested
Kimi (Moonshot)Cloudhttps://api.moonshot.ai/v1OpenAI-compatible · generic MoonshotBYO key
MiniMaxCloudhttps://api.minimaxi.com/anthropicMessages-shaped · Anthropic-styletested
OpenRouterCloudhttps://openrouter.ai/api/v1OpenAI-compatible · multi-upstream catalogBYO key
Bailian / DashScopeCloudhttps://dashscope.aliyuncs.com/compatible-mode/v1OpenAI-compatible · Alibaba-hosted modelsBYO key
OpenAI-compatibleCloudhttps://api.openai.com/v1OpenAI-compatible · any /v1 vendorBYO key

CLI configuration

Choose the executor. Keep the project truth stable.

The OwlCoda CLI can route to local or cloud models and expose real tools. Changing a model or session does not rewrite the project goal, accepted evidence, or delivery state.

  1. 01
    Print URL

    owlcoda ui prints a one-shot localhost URL bound to a session token. Open it manually, or pipe it into your launcher.

  2. 02
    Open browser

    owlcoda ui --open-browser hands the URL straight to your default browser.

  3. 03
    Configure

    Add a provider, paste an API key, run a test connection. The page closes itself when you’re done.

$ owlcoda ui
$ owlcoda ui --open-browser
$ owlcoda admin

Catalog, aliases, and orphan cleanup are tucked under Diagnostics. The first-run path stays calm.

owlcoda admin · providers · kimi-coding
01Wire
02Test
03Save

PROVIDER · KIMI CODING

ProviderKimi Coding
Endpointhttps://api.kimi.com/coding/v1
Backend modelkimi-for-coding
API keysk-•••••••••••••••
Wire formatprovider-native
Test200 OK · 412 ms

No port hardcoded. The URL changes per launch and never leaves your machine.

Coda

A coda is not just an ending. It is the closing passage that leaves the next starting point.

The project continues when the current agent stops.

OwlCoda keeps project truth in durable artifacts. A new agent reads the goal, current gap, candidate, evidence, decisions, and next action instead of reverse-engineering an old conversation.

$ npx owlrunkit project status --workspace . # recover the next action.

.owlcoda/runkit/ · project truthcurrent · inspectable
  • .owlcoda/runkit/project/definition.jsongoal · WorkItems · acceptance
  • .owlcoda/runkit/project/events/assignments · decisions · handoffs
  • .owlcoda/runkit/runtime-executions/execution evidence · receipts
  • .owlcoda/runkit/project/statuscurrent gap · owner · next action
  • SourceCandidate / DeliveryPacketexact candidate identity
  • DeliveryLifecycleintegration · deploy · live readback · acceptance

Control posture

Execution can change. Authority does not silently follow it.

  • Project truthThe current goal, candidate, evidence, and next action belong to the project — not to one executor transcript.
  • Explicit authorityGit, release, deployment, production, money, and business actions remain separate permissions.
  • InspectabilityFailures, handoffs, rework, recovery, and delivery stages remain visible instead of being rewritten as success.
  • Data controlExecution data stays in the environment and providers you choose; optional collection remains opt-in.

One public brand

One OwlCoda. Two product surfaces.

Users meet one brand. Delivery control is the primary product surface; the CLI is the auxiliary execution surface. The underlying repositories remain independently inspectable.

  • OwlCoda · Delivery Control ↗primary · powered by OwlRunKit 0.23

    Owns project continuity, candidate identity, verification, handoff, rework, recovery, and delivery lifecycle truth.

  • OwlCoda · CLI ↗auxiliary execution · 0.18

    Connects models and tools to execution, preserves sessions and evidence, and returns results to delivery control.