omadia
v0.18.017open

omadia

An Agentic OS.

:A coordinated team of specialised AI agents. Privacy-first, plugin-driven, no model lock-in. On your infrastructure.

reasoningverifierprivacymemorychannelOMADIAorchestrator

Open source (MIT) · GDPR-aware · self-hostable · Built in Germany

how it works

From models to teams.

An orchestrator dispatches each request to specialised agents. Each agent composes from pluggable capabilities. The privacy proxy sits in front of every public LLM and tokenises sensitive data reversibly. Receipts make every turn auditable.

Channel in

An end user (Teams, web, your own surface) sends a message. The channel SDK normalises it for the orchestrator.

Orchestrator routes

Picks the agent best fit for the task and dispatches to specialised Sub-Agents (accounting, HR, Confluence playbooks) when the task fits. Agents can hand off to one another.

Capabilities run

The agent calls capabilities (reasoning, memory, render, retrieval), each backed by a swappable provider.

Privacy + receipt

Sensitive data is tokenised before any public-LLM call. The answer is verified, then a receipt rides back with the response.

The difference

Agent frameworks give you parts. omadia gives you a system that holds up in an audit.

:Most agent stacks are libraries you assemble yourself. The hard parts stay your problem. Privacy, proof, lock-in, supply chain.

Without omadia

With omadia

You send prompts to an external model and hope nothing sensitive left the building. In a review, you cannot prove it.

Audit receipt

The privacy-proxy removes and reinserts sensitive data reversibly, and writes a receipt for every call. You hand the auditor receipts, not assurances.

The model makes things up with full confidence. Catching it before it ships is on you.

Hallucination verifier

A dedicated verifier checks every answer for drift before it reaches the user.

Your stack is welded to one provider's SDK and its pricing.

No model lock-in

Capability and provider are separate. Swap Claude for a local model in config, without rewriting agents.

Plugins arrive as npm packages, arbitrary code from a supply chain nobody audited.

Signed ZIP plugins

Plugins are signed ZIPs with a manifest. Install verifies the signature, so you know exactly what runs.

Want the widest set of community integrations today? A DIY framework still wins on raw count. omadia is for teams who have to answer to a security review.

privacy

Privacy that proves itself.

Sensitive values are tokenised reversibly, and every turn leaves a verifiable receipt.

Sensitive data never leaves your infrastructure in clear text. Before every public LLM call, the privacy proxy substitutes identifiers with tokens that resolve only locally. Every turn ends with a cryptographically auditable receipt: which tokens, which provider, which detectors. Your DPO gets concrete answers.

PRIVACY RECEIPT
rcpt-9a3f-d1e2 · 2026-05-09T08:42:11Z

Inbound (operator)

Customer Anna Schmidt asked about her order #ord-7421-A.

Outbound to LLM

Customer ⟨person-7a3f⟩ asked about her order ⟨order-id-91b2⟩.

Provider
anthropic / claude-4.7-opus
Detectors
person · iban · de-personalausweis · email
Status
redacted

Detected today (full stack)

Email · IBAN · Phone · Credit-Card (Luhn) · DE-Personalausweis · DE-Steuer-ID · Names (PERSON, NER) · Addresses · Organizations · custom regex

the builder

We built an agent that builds agents.

It writes the TypeScript, runs the typechecker, fixes its own lint errors, smoke-tests at runtime, and only ships when every pass is green. Output is a signed ZIP plugin, same install surface as any other.

Open the Builder
speccodegentypechecksmokeinstall

five gated passes · self-repair loop · signed-ZIP output

capabilities

Pluggable per capability.

Swap the LLM, the memory layer or the renderer without touching code.

Capabilities are versioned contracts. Providers are plugins that fulfil them. Today: Anthropic for chat, Ollama (self) for embeddings, three detector plugins for privacy. The matrix below shows where every other provider lands.

Capability × provider · roadmap
  • today
  • end of Q2 2026
  • not planned
capabilityClaudeGPT-5GeminiLlama (self)
LLM reasoningQ2Q2Q2
LLM codeQ2Q2Q2
EmbeddingQ2Q2
VisionQ2Q2Q2
Tool useQ2Q2Q2

plugins

Everything ships as a plugin.

Functionality ships as a signed ZIP plus manifest. No npm discovery, no supply-chain risk. Drop it in the Plugin Store, review the manifest, enable. Uninstall is clean, no shadow dependencies. Build your own plugin against the SDK in an afternoon.

The Plugin Store lives at /store in the operator UI.

01Plugin Store

Drop a .zip plugin here

or click to browse · signature verified on upload

Recently installed

  • installed

    teams-channel

    channel

  • installed

    privacy-guard

    extension

  • installed

    odoo-integration

    integration

proof · artifacts

Every claim here has an artifact.

The receipt, tokenisation and integrity check below come straight from the running code. Each one takes minutes to reproduce on your own machine.

privacy receipt
{
  "datasetsInterned": 1,
  "fieldsMasked": 1,
  "fieldsCleartext": 3,
  "verbsExecuted": ["sort", "top_n"],
  "pseudonymProjectionUsed": false
}

Emitted every agent turn. Counts what was interned, masked, and what left in clear. Your DPO audits the facts.

plugin-api · PrivacyReceipt

tokenisation
# operator input — stays on-prem
"Top customer: Anna Schmidt — 32 orders this quarter."

# payload sent to the LLM
"Top customer: [masked] — 32 orders this quarter."
  └─ interned as dataset ds_3f9a · resolves only locally

PII is interned behind a dataset handle that resolves only on-prem. The model only ever sees [masked].

privacy proxy · on-wire payload

plugin integrity
registry install · harness-plugin-web-search@1.0.0
  index      sha256:9f2c…b417   pinned in registry index
  artifact   download host-pinned to registry origin
  verify     sha256(zip) == pinned        ✓
  result     installed · capabilities reviewed first

  on mismatch → RegistryError: registry.sha256_mismatch

Every registry install is checked against a SHA-256 pinned in the index, from a host-pinned origin. A tampered artifact fails closed.

plugins · registryClient (sha256 pin)

Every artifact above comes straight from the OSS code.

integrations

Where omadia plugs in.

Pre-built connectors for the systems your team already runs on.

integration

Atlassian Confluence

confluence.client

Read-only, space-scoped REST API plus a Confluence-Playbook Sub-Agent and proactive page-entity sync into the knowledge graph.

  • Confluence-Playbook Sub-Agent (built-in)
  • Page-entity sync to KG
  • confluence.toolkit services
integration

Microsoft 365

m365.graph

Shared Microsoft Graph layer (Bot-Framework App-Registration). Powers Teams attachments, roster, and the calendar tools.

  • find_free_slots calendar tool
  • book_meeting calendar tool
  • Roster + delegated OBO-exchange
integration

Odoo

odoo.client

Read-only JSON-RPC plus a native enrich_company tool with OpenRegister + NorthData backed Handelsregister enrichment.

  • enrich_company native tool
  • Handelsregister via OpenRegister + NorthData
  • Read-only sync into KG

day one

Ten things you get on day one.

Everything here is included.

01

Privacy receipts

Per-turn audit log your DPO can read.

02

Multi-provider

Anthropic today. OpenAI, Gemini, self-host on the roadmap.

03

Diagrams

Mermaid · Graphviz · PlantUML · Vega-Lite via Kroki, signed PNG URLs.

04

Knowledge graph

Three retrieval paths on Neon + pgvector.

05

Hallucination verifier

Catches drift before it ships.

06

Channel SDK

Teams, Telegram, web, your own surface.

07

Bilingual UI

EN/DE first-class with a parity gate in CI.

08

Onboarding wizard

From empty workspace to working agent in under 15 minutes.

09

Quality Guard

Sycophancy levels and boundary presets at the response edge.

10

Web Search

Tavily / Brave with structured citations on every claim.

kemia · persona engine

Your agent, in your team's voice.

Twelve sliders for personality. Four boundary categories for what it must refuse. A sycophancy level your auditor can read. Snapshot the whole persona to a content-addressed ZIP and import it on another deployment. Identical behaviour by hash, no copy-paste prompts.

12 axes · default 50 = neutral · morph = two demo personas

for the eu operator

Built for the EU operator.

Bilingual from day one (EN/DE). Self-hostable on your infrastructure. GDPR-aware by design. Every provider call, every token substitution, every tool invocation lands in the audit log. Concrete answers for your DPO. Signed receipts for your auditor.

compliance · pilot ready
  • GDPR-aware design
  • Self-hostable on your infra
  • Per-turn audit trail
  • EN + DE first-class
  • MIT-licensed core
  • EU-hosted on request

stack

Built on open source.

Open-source components we run in production ourselves. License-clean, replaceable at every layer.

Platform
  • next.js 16
  • typescript 5
  • react 19
  • neon · pgvector
  • fly.io
  • tigris
  • kroki
  • ollama
MIT-licensed core · github.com/byte5ai/omadiagithub.com/byte5ai/omadia
15 min

From empty workspace to a working agent.

Onboarding wizard and a Docker-Compose quickstart. A promise with a number, not an adjective.

pilot

Ready for a pilot?

:We stand omadia up with you in two weeks and put the first agent in your team into production.