Kodakoda
Kodakoda

Many agents. Many providers.
One harness.

Koda is the open-source harness for multi-agent, multi-provider systems. A control plane you actually use, durable state by default, and every runtime action inspectable from day one.

Install Kodanpx@openkodaai/koda@latestinstall
Or read the documentation

The platform, day one.

Everything you need to run agents in production — configured from the dashboard, not six config files.

Control-plane-first configuration

Configure providers, agents, secrets, and access through the dashboard. No hand-maintained env files. First-run setup happens where operators already work.

Learn more
config / providers
synced
anthropic
claude-opus-4-7
openai
gpt-5
google
gemini-3-pro
ollama
local/llama3.3

Multi-agent, multi-provider orchestration

Claude, Codex, Gemini, Ollama — each agent picks the models and tools that fit its role. Isolated runtime state, parallel by default.

Learn more
koda / runtime
Atlas
anthropic/claude-opus-4-7
Live
Scout
google/gemini-3-pro
Live
Pebble
anthropic/claude-haiku-4-5
Queued
Otto
openai/gpt-5
Done

Knowledge & memory grounding

Retrieval, episodic memory, evidence sourcing, and durable context assembly — so every agent acts on what's actually true.

Learn more
/retrieve q1 outage
incidents/2026-q1-outage.md0.92
runbooks/postgres-replica-recovery.md0.78
skills/incident-triage.md0.64

Skills as agent abilities

Stored expert prompts exposed via /skill. Reusable, inspectable, and shareable across agents — your best patterns without the copy-paste.

Learn more
/skills catalog
4 shared
/rcaRoot-cause analysis
142 runs
/summarize-prPR summary from diff
88 runs
/db-migration-planSchema migration draft
31 runs
/incident-triageClassify + route incident
67 runs

Durable artifact processing

Postgres for state, S3-compatible object storage for binaries. Documents, media, and derived evidence kept with inspection chains intact.

Learn more
artifacts / run-2417
PGS3
outage-screenshot.png
2.4 MB · S3
postmortem-q1.pdf
488 KB · S3
evidence-chain.json
12 KB · PG

Runtime visibility

Health, setup, dashboards, and OpenAPI-backed surfaces. Every tool call, provider response, and memory write is auditable by default.

Learn more
openapi / control-plane
99.94%
GET/api/runtime/agents14ms
POST/api/runtime/tasks42ms
GET/api/control-plane/health8ms
GET/api/runtime/tasks/:id21ms
Integrations

Speaks the protocols you already use.

Anthropic
OpenAI
Gemini
Ollama
PostgreSQL
Docker
AWS S3
GitHub
Vercel
Cloudflare
Slack
Linear
Anthropic
OpenAI
Gemini
Ollama
PostgreSQL
Docker
AWS S3
GitHub
Vercel
Cloudflare
Slack
Linear
Anthropic
OpenAI
Gemini
Ollama
PostgreSQL
Docker
AWS S3
GitHub
Vercel
Cloudflare
Slack
Linear
Atlas
anthropic/claude-opus-4-7
Otto
openai/gpt-5
Scout
google/gemini-3-pro
Pebble
anthropic/claude-haiku-4-5
Sage
anthropic/claude-opus-4-7
Whiskers
ollama/llama3.3
Marlowe
openai/gpt-5
Beacon
google/gemini-3-pro
Echo
anthropic/claude-opus-4-7
Sprout
anthropic/claude-haiku-4-5
Atlas
anthropic/claude-opus-4-7
Otto
openai/gpt-5
Scout
google/gemini-3-pro
Pebble
anthropic/claude-haiku-4-5
Sage
anthropic/claude-opus-4-7
Whiskers
ollama/llama3.3
Marlowe
openai/gpt-5
Beacon
google/gemini-3-pro
Echo
anthropic/claude-opus-4-7
Sprout
anthropic/claude-haiku-4-5
Open foundations

Production-friendly.
Operator-first.

Koda treats agent configuration the way operators treat production systems — through a control plane, with inspection, audit trails, and durable state. Programmable where you want it. Steerable where you need it.

MIT-licensedSelf-hostableMulti-providerPostgres + S3

No black boxes. No hidden state.

Running a multi-agent system surfaces a hundred operational questions. Koda answers them in its design — not as afterthoughts wrapped around a hardcoded prompt loop.

How do I prevent agent runaway?Which model fits this task?Can I switch providers without rewrites?Where is each agent's state stored?How do I prevent agent runaway?Which model fits this task?Can I switch providers without rewrites?Where is each agent's state stored?How do I prevent agent runaway?Which model fits this task?Can I switch providers without rewrites?Where is each agent's state stored?How do I prevent agent runaway?Which model fits this task?Can I switch providers without rewrites?Where is each agent's state stored?
How do I roll back a bad action?What did the agent see at step 3?Can I run this entirely on my own infra?How do I share a Skill across agents?How do I roll back a bad action?What did the agent see at step 3?Can I run this entirely on my own infra?How do I share a Skill across agents?How do I roll back a bad action?What did the agent see at step 3?Can I run this entirely on my own infra?How do I share a Skill across agents?How do I roll back a bad action?What did the agent see at step 3?Can I run this entirely on my own infra?How do I share a Skill across agents?
Why did the agent retry?What's the audit trail for this run?Where do agents pull memory from?Can I chain agents and inspect each step?Why did the agent retry?What's the audit trail for this run?Where do agents pull memory from?Can I chain agents and inspect each step?Why did the agent retry?What's the audit trail for this run?Where do agents pull memory from?Can I chain agents and inspect each step?Why did the agent retry?What's the audit trail for this run?Where do agents pull memory from?Can I chain agents and inspect each step?

Programmable, not prescriptive

No prescribed agent persona, no locked task domain. Configure providers, agents, and policies the way your system actually needs them.

Production-grade by design

Postgres for durable state. S3-compatible artifact storage. OpenAPI-backed public surfaces. Built for operators, not hobby projects.

Inspect every action

Every tool call, retrieval hit, and provider response is logged with full context. The runtime state is never hidden behind the prompt.

Yours to fork, yours to run

MIT-licensed. Docker-first. Self-host on a single VPS or scale across nodes. No managed service, no vendor path dependency.

Start with the platform.

One command to bring up the stack — Postgres, object storage, runtime, and dashboard. Configure providers, agents, and skills from there.

Install Kodanpx@openkodaai/koda@latestinstall
Or read the documentation