Openloomi
OpenLoomi is an open-source, local-first AI coworker whose desktop attention agent turns your screens and connected apps into lasting work memory.
OpenLoomi makes a real design bet: memory plus approval-gated action, not another chat wrapper. If your work lives in Slack, Gmail, and WhatsApp and you care where that data sits, the Apache 2.0 codebase, the 0.8.0 unified im_reply workflow, and local-first AES-256 handling make it worth cloning this week. The desktop attention agent and the Approve-before-execute Loop are genuinely differentiated from recall-only tools like Rewind. If your team runs on GitHub, Jira, or Linear, wait — those connectors are still labeled 'Soon' on the homepage, and the plug-in path through Composio is how you'd bridge them today.
Verified 1d ago · liveness 62/100 · cite: rightaichoice.com/tools/openloomi
- Developers and builders who want local-first work memory wired into Claude Code or Codex
- Privacy-conscious technical professionals who won't hand work context to a cloud vendor
- Small engineering teams comfortable self-hosting an Apache 2.0 codebase from GitHub
- Founders and knowledge workers drowning in Slack, Gmail, and WhatsApp who want an AI that remembers across sessions
- Non-technical users who want a zero-setup cloud assistant; the CLI and self-hosting are part of the deal
- Teams whose workflow centers on GitHub, Jira, Linear, or Notion; those connectors are still marked 'Soon'
- Buyers who need a completed SOC 2 report today; compliance is listed as in progress
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip OpenLoomi if your team's work lives primarily in GitHub, Jira, or Linear — those connectors are labeled 'Soon' on the homepage and you would need the Composio path or a different tool today.
OpenLoomi is distributed as Apache 2.0 source you clone from GitHub, so your effective cost is machine time, an AI provider subscription for the models you point it at, and the engineering hours to run the desktop app or plugin yourself. Teams that want a managed SaaS knowledge base instead — Glean, for instance — pay per seat and get none of the local-first guarantees OpenLoomi makes.
In short
Openloomi — OpenLoomi is an open-source, local-first AI coworker whose desktop attention agent turns your screens and connected apps into lasting work memory. Best for Developers and builders who want local-first work memory wired into Claude Code or Codex, Privacy-conscious technical professionals who won't hand work context to a cloud vendor, Small engineering teams comfortable self-hosting an Apache 2.0 codebase from GitHub. Free to use.
What's new in Openloomi
Checked yesterdayAcross the latest 2 updates: 2 changelog entries.
OpenLoomi 0.8.0 adds unified im_reply decision type
OpenLoomi 0.8.0 ships a unified im_reply decision type, graph-driven memory consolidation, and hardened plugin setup.
OpenLoomi 0.7.0 introduces AI coworker experience and Loomi pet
OpenLoomi 0.7.0 adds an AI coworker experience, the Loomi pet, integrated Loop, and memory intelligence.
What people actually say about Openloomi — is it worth it?
We ran a structured research pass across product reviews, community discussions, and post-purchase forum threads to surface the patterns vendors won't publish themselves. Below: the recurring strengths, the hidden costs people mention most, and the cohort that consistently regrets adopting this tool.
2 mentions across 1 source (Hacker News) · researched Jul 3, 2026.
Average across the 1 source that answered — each source counts once, not each post.
- +Persistent context graph remembers work history across sessions.
- +Local-first architecture keeps raw data on your device.
- +Open source under Apache 2.0 allows full customization.
- +Proactive task execution with mandatory user approval.
- +Integrates Slack, email, calendar, and docs into one view.
- −Very limited community feedback and real-world testing.
- −Setup and integration may be complex and time-consuming.
- −Proactive AI may generate false positives or irrelevant actions.
- −Lack of independent privacy audits raises trust questions.
- −No mobile app or desktop client mentioned.
- • Self-hosted Enterprise may require additional infrastructure costs
- • Potential add-on fees for premium integrations
Viability Score
How well maintained and how widely used is Openloomi? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Desktop attention agent reads authorized screen content to build work context
- Holistic context graph links people, projects, decisions, and follow-ups
- Tiered memory storage with a forgetting engine and semantic search
- Proactive Loop engine turns scattered signals into approval-gated decision cards
- Human-in-the-loop Approve step before every executed action
- Draft replies, schedule follow-ups, log updates, delegate, and close tickets
- Unified im_reply decision type for messaging workflows (v0.8.0)
- Graph-driven memory consolidation and memory relations (v0.8.0)
- AI coworker experience with the Loomi pet (v0.7.0)
- Connectors auto-fetch commits, issues, emails, and docs into one context graph
- Claude Code plugin with /openloomi:* commands
- Codex plugin for persistent work memory
- Agent runtimes: Claude Code, Codex, OpenCode, Hermes, OpenClaw
- Automation: schedule jobs, react to events, trigger cross-app flows
- Library: upload documents and query a local knowledge base
About Openloomi
OpenLoomi is an open-source, local-first AI coworker for people who are tired of re-explaining their job to a chatbot every morning. A desktop attention agent reads the screen content and connected apps you authorize, then builds a holistic context graph linking your people, projects, decisions, and follow-ups instead of treating each session as a blank slate. The vendor's own framing is blunt: the bottleneck isn't model capability, it's AI awareness of your work. The Loop engine is what separates it from a plain memory layer. Instead of stopping at recall, OpenLoomi drafts replies, schedules follow-ups, logs updates, delegates, and closes tickets — but every action sits behind a human-in-the-loop Approve step, so nothing fires without you. Release 0.7.0 added the AI coworker experience and the Loomi pet alongside an integrated Loop, and 0.8.0 shipped a unified im_reply decision type for messaging workflows plus graph-driven memory consolidation and hardened plugin setup. Documented live connectors cover Slack, Telegram, Discord, WhatsApp, Gmail, Outlook, iMessage, Lark/Feishu, QQ, X, and Google Docs, plus 1000+ apps via Composio; the homepage separately labels HubSpot, Outlook Calendar, Google Drive, GitHub, Google Calendar, LinkedIn, Instagram, Facebook Messenger, Microsoft Teams, Notion, Asana, Jira, and Linear as 'Soon' — worth reading literally before you commit. Agent runtimes span Claude Code, Codex, OpenCode, Hermes, and OpenClaw, with Claude Code and Codex plugins for a one-command hookup. The privacy story is unusually specific: Apache 2.0 source at github.com/melandlabs/openloomi, raw messages and files staying on your machine, AES-256 encryption on authorized data processed inside a hardware-isolated trusted execution environment, tamper-proof audit logs, and a zero-training pledge the company says is legally guaranteed. SOC 2 is listed as in progress, not complete. It fits developers and technical founders who already work inside Claude Code or Codex and prefer a git clone to a cloud subscription; anyone wanting a zero-setup hosted assistant should look elsewhere.
Behind the Verdict
OpenLoomi's core claim is that the bottleneck isn't model capability but AI awareness of your work — and the product is built around that thesis rather than around a bigger model. A desktop attention agent reads the screen content and connected apps you authorize, then a tiered memory store with a forgetting engine and semantic search keeps people, projects, decisions, and follow-ups linked across sessions. The docs describe a LoCoMo, LongMemEval, and CL-bench benchmark page for the memory system, which is a more honest way to sell memory than a vague 'remembers everything' claim. Where it separates from recall-only competitors like Rewind or Microsoft Recall is the Loop: proactive execution engine that drafts replies, schedules follow-ups, logs updates, delegates, and closes tickets, with every action gated behind an Approve step. The 0.8.0 release added a unified im_reply decision type for messaging workflows plus graph-driven memory consolidation; 0.7.0 brought the AI coworker experience, the Loomi pet, and the integrated Loop. Practically, that means you get decision cards rather than a passive transcript archive. The integrations picture is the honest weak point. The homepage's live list is Slack, Telegram, Discord, WhatsApp, Gmail, Outlook, iMessage, Lark/Feishu, QQ, X, and Google Docs, plus Composio for 1000+ apps; HubSpot, GitHub, Google Calendar, Outlook Calendar, Google Drive, Microsoft Teams, Notion, Asana, Jira, and Linear are all still labeled 'Soon'. Developer shops that live in GitHub and Linear will need to bridge those through Composio rather than wait. Deployment is a clone-and-run story: Apache 2.0 source at github.com/melandlabs/openloomi, local-first storage, AES-256 encryption inside a hardware-isolated trusted execution environment, tamper-proof audit logs, and a zero-training pledge the company describes as legally guaranteed. SOC 2 is explicitly in progress rather than complete, so regulated buyers have nothing to request today. If you already live in Claude Code or Codex, the one-command plugin path gives you persistent work memory inside the coding surface you already use; the docs say most readers finish a first Approve within 30 minutes that way. If you want a zero-setup cloud assistant, this is the wrong tool.
Researching Openloomi? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Openloomi actually fits — and what changes day-one when you adopt it.
Install the OpenLoomi plugin with the one-command setup, authorize Gmail and Slack as connectors, and let the Loop run while you work in the CLI; the plugin gives your coding agent persistent memory of people, projects, and decisions.
Outcome: Your coding agent references past decisions and open follow-ups instead of starting from scratch, and Loop surfaces decision cards you approve inside the same surface.
Install the desktop app, authorize Gmail and Slack first (the docs call those the most useful starting pair), then let the attention agent build the context graph for an afternoon; review the decision cards the Loop produces for drafts, follow-ups, and logging.
Outcome: Context-aware drafts in your unified inbox and tracked follow-ups, with every send and log gated behind an Approve step.
Route inbound email and messaging through connectors, let the Loop consolidate the thread history into a graph, and approve the drafted resolution or delegation before it fires.
Outcome: Tickets closed with a summarized resolution and handoffs delegated with pending-action tracking, without any message sent without your sign-off.
Use Cases
- Draft replies to emails or messages with full context of past conversations and decisions.
- Schedule follow-ups automatically after meetings or missed commitments.
- Log updates to projects or tickets based on discussion history.
- Delegate requests to team members while tracking pending actions.
- Close support tickets by summarizing resolution and getting your approval.
- Use from inside Claude Code or Codex CLI to give your coding agent persistent work memory.
- Automate recurring workflows like daily syncs or content drafting via scheduled jobs and event triggers.
- Run a unified inbox that auto-fetches mail and generates context-aware drafts.
Models Under the Hood
as of 2026-10-18
Limitations
- OpenLoomi is an early, self-hosted product.
- Many connectors are labeled 'Soon' on the homepage: HubSpot, Outlook Calendar, Google Drive, GitHub, Google Calendar, LinkedIn, Instagram, Facebook Messenger, Microsoft Teams, Notion, Asana, Jira, and Linear.
- SOC 2 compliance is in progress rather than complete, so regulated buyers have no report to request today.
- Setup assumes CLI comfort: install the desktop app or the Claude Code / Codex plugin, configure an AI provider, authorize connectors, and wire the runtimes.
- Help comes from the docs and a community Discord — there is no vendor support SLA.
- Model provider configuration is your responsibility: Anthropic-compatible providers with RPM support and embedding providers are documented in the reference section.
as of 2026-10-10
Verification history
We have re-verified Openloomi 9 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-checked, vendor evidence unchanged
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-checked, vendor evidence unchanged
Showing the 6 most recent of 9 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Openloomi tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Self-Hosted
$0/mo
Ideal for
Developers and technical founders comfortable cloning an Apache 2.0 repo and running the desktop app or plugin on their own machine
What this tier adds
Starting (and only published) entry point: the full product via git clone, with local-first storage, the attention agent, the Loop, and the documented connectors
Where the pricing makes sense
The company stage and team size where Openloomi's pricing actually pencils out — and where peers do it cheaper.
OpenLoomi is distributed as Apache 2.0 source you clone from GitHub, so your effective cost is machine time, an AI provider subscription for the models you point it at, and the engineering hours to run the desktop app or plugin yourself. Teams that want a managed SaaS knowledge base instead — Glean, for instance — pay per seat and get none of the local-first guarantees OpenLoomi makes.
Setup time & first value
How long it actually takes to get something useful out of Openloomi — broken out by persona, not the marketing-page minute.
The docs claim most readers finish a first Approve within 30 minutes. Desktop path: install the app, sign in, configure an AI provider, authorize one connector (Gmail or Slack is the most useful), and let Loop run for an afternoon. Plugin path: if you already live in Claude Code or Codex, the one-command plugin setup skips the desktop entirely. Expect longer if you are self-hosting and wiring
Switching to or from Openloomi
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From a cloud chat assistant like ChatGPT or Claude: clone the Apache 2.0 repo, run the desktop app locally, and let the attention agent rebuild your people/project/decision context from your connected apps.
- →From Rewind or Microsoft Recall: install OpenLoomi, authorize the same apps, and gain the Loop's approval-gated actions on top of the recall you already had.
- →From a manual notes or second-brain system: upload your documents to the Library and let memory consolidation and semantic search replace the manual linking.
- →From no assistant at all: start with the Getting Started doc, authorize Gmail or Slack, and run Loop for one afternoon to reach a first Approve.
- ↗To a managed AI knowledge base like Glean: export your Library documents and manually recreate the people/project relationships Glean needs, since the context graph format does not port.
- ↗To Claude Code or Codex memory plugins: if you only used OpenLoomi for coding-session memory, swap to a plugin-native memory layer and drop the desktop app.
- ↗To Microsoft Copilot or Google Gemini in-workspace: acceptable if your work is already inside Microsoft 365 or Google Workspace and you don't need the local-first guarantees.
Integrations
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “Openloomi”, and we withheld 6: 6 could not be judged, because “Openloomi” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Openloomi.
Official links
Tools that pair well with Openloomi
Common stack mates teams adopt alongside Openloomi, with the specific reason each pairing earns its keep.
Featured Head-to-Head Comparisons
Openloomi vs Locus Robotics
Locus Robotics is the clear choice if you need physical warehouse automation to boost picking productivity 2-3x in high-volume fulfillment centers. Openloomi, on the other hand, is ideal if you want an open-source, local-first AI coworker that persists context across your digital tools and automates tasks with strong privacy guarantees. They solve completely different problems, so pick based on whether your pain is physical or digital workflow automation.
Openloomi vs Truleo
If you're in law enforcement and drowning in siloed data (jail calls, body cameras, RMS), Truleo is your only choice. For everyone else—especially privacy-conscious team leaders or developers—Openloomi's open-source, local-first approach with persistent work memory is far more flexible and cost-effective. No contest: pick the tool built for your world.
Openloomi vs Presto Voice
Presto Voice is purpose-built for QSR chains needing proven drive-thru automation with upselling, as validated by its Dairy Queen partnership. Openloomi offers a flexible, open-source AI coworker for teams that want persistent memory and local control, but requires technical setup. Choose Presto for specialized enterprise voice AI; choose Openloomi for general-purpose, privacy-first workflow automation.
Alternatives to Openloomi
View all
Goose by Block
Goose is a local open-source AI agent for code, workflows, and automation via MCP.
Popular in Productivity
Genspark
Genspark is an agentic AI workspace that turns one prompt into cited answers, slide decks, dashboards and live web apps.
Poke (Interaction Co.)
Poke is an AI personal assistant that lives in your texts on Apple Messages, WhatsApp, Telegram and RCS
Frequently Asked Questions
Best-of guides
Used Openloomi? Help shape our editorial sentiment research.