Omnara

Omnara

Unclaimed verified 17 nov 2026
[  score · 34  ]

Mobile & Voice Interface for Claude Code & Codex

Pricing: Freemium - $20/month Company: Omnara Founded: 2025 Last verified: 2026-11-17
Visit Website
Updated

TL;DR

Omnara is a mobile, web, and voice command center and open-source orchestration platform for AI coding agents such as Claude Code and Codex. It enables developers to start coding jobs on local or remote machines and monitor, steer, or approve tool actions hands-free from anywhere. Its primary differentiator is seamless cross-device session continuity with bidirectional voice and push-approval controls.

What Users Actually Pay

No user-reported pricing yet.

Our Take

Omnara tackles a widespread friction point in agentic software engineering: being anchored to a terminal waiting for an AI agent to request permission or ask clarification questions. By bridging local CLI daemons (`omnarad`) with mobile push notifications, diff viewers, and two-way voice input, Omnara allows developers to treat coding agents as asynchronous collaborators rather than fragile terminal processes. The tool has matured from a mobile client into an open-source, Apache 2.0-licensed agent orchestration backend. Its strength lies in model and platform independence, offering first-party support for Model Context Protocol (MCP) servers, multi-model routing (Anthropic, OpenAI, OpenRouter, LiteLLM), and self-hosting options backed by PostgreSQL and Docker. However, limitations exist for enterprise environments. Running agents across machines without self-hosting requires streaming diffs and terminal states to Omnara's cloud coordination layer, and migrating mid-session from local setups to cloud sandboxes can require managing environment parity manually. Additionally, it lacks formal enterprise compliance certifications like SOC2. Omnara is best suited for solo software engineers, fast-moving product teams, and platform developers heavily reliant on Claude Code or Codex who want to review PRs, grant execution permissions, or kick off long-running refactors while away from their desks.

Pros

  • + Real-time mobile push notifications for instant tool approvals, prompt responses, and diff inspection.
  • + Conversational two-way voice interface allowing hands-free task delegation and code reviews.
  • + Open-source (Apache 2.0) architecture supporting private self-hosting via Docker and PostgreSQL.
  • + First-party Model Context Protocol (MCP) registry integration and native support for GitHub, Slack, and Discord.

Cons

  • - No native SOC2 or enterprise audit logging in managed tiers without self-hosting.
  • - Occasional dependency/environment parity gaps when moving running tasks from local machines into remote cloud containers.
  • - Rapidly shifting product boundaries from a mobile CLI companion to a full agent backend system.

Agent Readiness

71/100

Omnara offers robust programmatic and autonomous agent readiness, featuring an open-source REST/SSE API, official TypeScript SDK, Model Context Protocol (MCP) server support, and daemon workers. While it lacks pre-built no-code connectors for Zapier or Make, its native GitHub, Slack, and webhook integrations provide direct extensibility for automated developer workflows.

API Surface100
Public APIRESTWebSocketSSEFree TieropenApi
Protocol Support55
MCP (22 tools)
SDK Availability70
npm: @omnara/sdk (official)npm: omnara (official)npm: @ag-ui/omnara (official)npm: n8n-nodes-omnara (official)pypi: omnara (official)
Integration Ecosystem25
WebhooksModel Context Protocol (MCP)GitHub AppSlack BotDiscord BotPostgreSQLClaude Code CLIOpenAI Codex
Developer Experience85
Docs: goodSandboxVersioningChangelog

Last checked Nov 5, 2026

MCP Integrations

3 servers22 tools
Omnara
Omnaraomnara/omnara-mcp
smitheryRemoteHigh match

Omnara is the open-source managed agent platform (Apache 2.0, backed by Y Combinator, 2.9k GitHub stars). Its hosted MCP server gives your coding agent the whole Omnara API, so "build me an agent that..." becomes a running production agent without leaving your editor. What your coding agent can do: - Launch agents: write the config, pick any model (OpenAI, Anthropic, OpenRouter, or your own endpoint) and a machine (a managed sandbox or your own laptop or VM), and start it. - Work with running agents: send messages, steer them mid-task, read every event, cancel a turn, or archive the agent. - Handle approvals: see the tool calls and questions an agent is waiting on, and allow or deny them. - Schedule work: create cron triggers that launch a new agent from a profile or message an existing one. - Manage your workspace: profiles, model providers, machines, sandbox pools, secrets, and which projects can use each. Agents run on Omnara, not inside your chat. Every step is saved, so they kee

22 tools
  • grant_secrets_addCreate secret grant
  • grant_secrets_deleteDeletes only this project grant. The secret, its versions, the target project, and other grants are retained, and there are no dependency blockers.
  • grant_secrets_listItems are ordered by created_at descending, then id descending.
  • grant_skills_addCreate skill grant
  • grant_skills_deleteDeletes only this project grant. The skill, its revisions, the target project, and other grants are retained, and there are no dependency blockers.
  • grant_skills_listItems are ordered by created_at descending, then id descending.
  • machines_listItems are ordered by machine created_at descending, then machine id descending.
  • model_providers_catalogFetches the provider's live model catalog by validating the configured credential and probing the provider's /models endpoint. The call waits on the upstream provider, so it is slower than the other model provider config routes and can report failure as data.
  • model_providers_listList model provider configs with server-side filtering, sorting, and cursor pagination.
  • orgs_createCreates the organization and its local default resources atomically. An idempotency key determines the proposed organization ID; replaying it returns the existing organization. Without a key, each request is a fresh attempt. Optional default-provider provisioning begins after commit and retries asynchronously without failing organization creation.
  • orgs_listUsers get every organization they belong to; org API keys get the organization that issued them. Items are ordered by organization name ascending.
  • projects_listItems are ordered by created_at descending, then id descending.
  • secrets_createCreate secret
  • secrets_deleteDeletes the secret and permanently destroys all of its versions and project grants; the ciphertext cannot be recovered. Deletion is blocked while the secret is the credential of a model provider config, machine pool, or integration.
  • secrets_getGet secret through ownership authority
  • secrets_listItems are ordered by created_at descending, then id descending. Without an owner filter, returns the deduplicated union of organization secrets the caller may list, project secrets the caller may list, and the caller's own user secrets.
  • secrets_updateUpdate secret metadata
  • skills_deleteDeletes the skill together with all of its revisions and project grants; uploaded archive content is destroyed. No other resources are deleted. Deletion is blocked while an active agent's configuration references the skill.
  • skills_getGet skill through ownership authority
  • skills_listItems are ordered by created_at descending, then id descending. Without an owner filter, returns the deduplicated union of organization skills the caller may list, project skills the caller may list, and the caller's own user skills. Grants are managed separately and do not change ownership.
  • tools_catalogGet tool catalog
  • whoamiGet the authenticated user and their organizations
com.omnara/omnaracom.omnara/omnara
officialRemoteHigh match

Deploy durable AI agents with Omnara, an open-source managed-agent platform.

io.github.justjlee/omnarai-mcpio.github.justjlee/omnarai-mcp
officialRemote

Deliberation + live 5-model council divergence over the Omnarai multi-AI attributed corpus.

Last checked Oct 25, 2026

Screenshot

Omnara screenshot

[ features ]

Compliance & Security

Security certifications, compliance features, and access control capabilities.

SOC 2

SOC 2 Type I or Type II certification.

none
ISO 27001

ISO 27001 information security certification.

no
GDPR Tools

Built-in tools for GDPR compliance (data export, deletion, consent).

[  yes  ]
Audit Trail

Complete audit log of all data changes.

—
Role-Based Access Control

Granular permissions based on user roles.

—
SSO Support

Single Sign-On integration support.

none

Reviews

0 reviews
Write a Review

No reviews yet. Be the first to review Omnara!