N8n Claw is live · free · last verified 25 Oct 2026 · viability monitor
N8n Claw
Open-source, self-hosted AI agent on n8n: Telegram-first chat, adaptive memory, 70+ MCP skills, 100+ Expert Agents.
If you already run n8n and want an agent that lives in your own infrastructure, n8n-claw is one of the more complete open-source options: 70+ installable MCP skills, 100+ Expert Agents, a knowledge-graph memory layer, and a Chromium browser with Telegram-based 2FA round-trip that most chat-wrapper agents simply don't have. The trade-off is real — you own the server, the updates, and the troubleshooting, and there is no vendor support line. Compare it to OpenClaw for autonomous coding (n8n-claw delegates to it rather than replacing it) and to managed assistants if you'd rather not run infrastructure at all.
Verdict generatedVerified 5d ago · liveness 65/100 · cite: rightaichoice.com/tools/n8n-claw
- Developers who want a self-hosted AI agent on infrastructure they control
- Existing n8n users extending workflows with an autonomous agent
- SMEs automating task management and CRM work without per-seat SaaS fees
- Privacy-conscious power users avoiding cloud lock-in
- Non-technical users who want a fully managed SaaS assistant
- Teams that need a dedicated mobile app rather than Telegram or API access
- Organizations requiring vendor SLAs and enterprise support contracts
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 n8n-claw if you want a managed assistant with vendor support and no server to run — it is self-hosted by design and every path to first value starts with your own infrastructure.
Running the agent means paying for a server plus whatever your chosen LLM provider charges per token, since inference costs sit with your provider account rather than the project.
n8n-claw itself is an open-source, self-hosted project, so your recurring spend lands with your server host and your LLM provider rather than the vendor. For a solo developer on Ollama, that can be near zero marginal cost; for a team routing heavy traffic through Claude or GPT, inference becomes the dominant line. Compare against per-seat managed assistants where cost scales with headcount — the trade is engineering time instead of licences.
In short
N8n Claw — Open-source, self-hosted AI agent on n8n: Telegram-first chat, adaptive memory, 70+ MCP skills, 100+ Expert Agents. Best for Developers who want a self-hosted AI agent on infrastructure they control, Existing n8n users extending workflows with an autonomous agent, SMEs automating task management and CRM work without per-seat SaaS fees. Free to use.
What is N8n Claw?
N8n Claw — Open-source, self-hosted AI agent on n8n: Telegram-first chat, adaptive memory, 70+ MCP skills, 100+ Expert Agents. Best for Developers who want a self-hosted AI agent on infrastructure they control, Existing n8n users extending workflows with an autonomous agent, SMEs automating task management and CRM work without per-seat SaaS fees. Free to use.
n8n-claw is an open-source AI agent you run on your own server, built on top of n8n workflows. It positions itself as a direct alternative to OpenClaw. You talk to it through Telegram, Slack, Teams, or your own app via an HTTP webhook API, and it manages tasks, memory, reminders, and integrations in natural language. Its memory layer combines hybrid search (semantic + full-text + entity, fused with RRF), time decay, and a knowledge graph of people and companies; nightly pattern extraction turns past conversations into a behavior model of how you work. Tasks get priorities, due dates, and proactive nudges for overdue items, while recurring instructions like "check my emails every 15 minutes" or "daily briefing at 8am" are scheduled conversationally. Extensibility is the real differentiator: 19+ built-in tools, a skill catalog of 70+ pre-built MCP integrations (HubSpot, Salesforce, Zoho, Jira, Confluence, Asana, Stripe, Notion, Todoist, Home Assistant, Google Services, and more), and 100+ Expert Agents split across analytics, marketing, development, sales, creative, operations, product, and research. Web access comes in three layers — SearXNG search with no API key, Crawl4AI page reading, and a real Chromium browser that handles logins, form fills, and interactive 2FA through a Telegram round-trip. Setup is a repo clone plus one script with interactive prompts for your Telegram bot token, LLM provider, and agent personality; it supports Claude, GPT, Gemini, Mistral, DeepSeek, Ollama, OpenRouter, or any OpenAI-compatible endpoint. It fits developers and technical teams who want data ownership and deep n8n integration, and it does not fit anyone expecting a managed SaaS product with a polished mobile app or vendor SLAs.
Key features
- Chat via Telegram, Slack, Teams, or HTTP webhook API
- Adaptive RAG memory with hybrid search (semantic + full-text + entity, fused via RRF)
- Time decay so older context ranks lower in retrieval
- Knowledge graph of people and companies
- Nightly pattern extraction building a behavior model of how you work
- Custom personality: agent name, language, and casual/professional/custom tone
- Task management with priorities, due dates, and proactive overdue reminders
- One-time reminders, scheduled actions, and timed instructions in natural language
- Recurring actions like "check my emails every 15 minutes" or "daily briefing at 8am"
- Expert Agents: 3 included out of the box, expandable from a catalog of 100+
- MCP skill catalog with 70+ pre-built integrations installed by asking in chat
- Google Services OAuth2 for Gmail, Calendar, Analytics, and Ads, with the auth flow handled via Telegram
- Web search via SearXNG with no API key required
- Clean page reading via Crawl4AI
- Real Chromium browser automation for signups, form fills, logins, and click flows
Behind the verdict
n8n-claw's pitch is ownership over convenience, and the feature list backs that up rather than just asserting it. The memory architecture is the most interesting part: hybrid search combining semantic, full-text, and entity retrieval fused with RRF, plus time decay so older context fades, plus a knowledge graph that tracks people and companies. Nightly pattern extraction then converts conversation history into a behavior model, which is what makes the proactive task reminders feel less like a cron job and more like an assistant that noticed you keep pushing the same follow-up. That is a meaningfully deeper memory design than the flat vector store most self-hosted agents ship with. The skill system is the second strength. Instead of hardcoding every integration, n8n-claw lets you say "install the HubSpot skill" in chat and get contact creation, deal updates, and activity logging without touching config files. The catalog covers CRM (HubSpot, Salesforce, Zoho), project management (Jira, Confluence, Asana), payments (Stripe, Lexware), banking (KontoFlux read-only across 5,000+ German accounts), spreadsheets (Airtable, NocoDB), Notion, Todoist, Gmail and IMAP/SMTP, Google Calendar and CalDAV, Home Assistant, cloud storage (Google Drive, Nextcloud, Seafile), the Vexa meeting bot for live transcription, and generative media via Nano Banana Pro for images and Veo 3.1 for video. Expert Agents add 100+ specialists — 25 in development, 16 in marketing, 14 in operations, 9 in analytics — that the main agent delegates to when a task matches. The web layer earns its place too. SearXNG search needs no API key, Crawl4AI handles clean page reading, and a real Chromium session persists across calls so the agent can do newsletter signups, form fills, logins, and click flows — with interactive 2FA completed by asking you for the code in Telegram. That browser persistence is genuinely uncommon in this category and is often the difference between an agent that can only read the web and one that can actually operate it. Where it gets harder: everything above assumes you are comfortable running a server, managing n8n, and debugging when a workflow breaks. There is no managed cloud option and no official support, so your fallback is community resources and your own reading. The Telegram-first interface is excellent for quick capture — voice messages, photos of receipts, a location drop — but if your team lives in a browser dashboard, you'll be improvising one. LLM choice is broad and made at setup time, and you can pick Claude, GPT, Gemini, Mistral, DeepSeek, Ollama, OpenRouter, or any OpenAI-compatible endpoint, so you are not locked to one model vendor. Ten-minute setup is a fair claim if your server is already prepared and you have a Telegram bot token in hand; the clock starts after that.
Models under the hood
as of 2026-10-25
Limitations
- You host it yourself, which means you own server setup, updates, and scaling.
- The agent is built on n8n workflows, so non-developers face a real learning curve when something needs debugging.
- There is no managed cloud option and no official support channel, so troubleshooting leans on community resources.
- The interface is Telegram-first, with Slack, Teams, and HTTP webhook access available, but no dedicated mobile app or browser dashboard.
- Expert Agents ship with 3 enabled out of the box, so the full 100+ catalog requires expansion work on your side.
as of 2026-10-08
Researching N8n Claw? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Is N8n Claw still active?
N8n Claw is live · free · last verified 25 Oct 2026 · viability monitor.
Viability score
How well maintained and how widely used is N8n Claw? 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 →What do users say about N8n Claw?
What people actually say about N8n Claw — is it worth it?
Our community sentiment research written Jul 3, 2026 is more than 90 days old, so it is being re-reviewed against current data and is not shown.
How we verified this
Every listed fact about N8n Claw— pricing, features, links, status — is re-read from the vendor's own pages by our verification pipeline and each change is logged with a date; the last pass ran on 25 Oct 2026. Pages mix model-assisted synthesis with human review; see how we verify.
Verification history
We have re-verified N8n Claw 12 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.
- — changed switching out, hidden costs, who it suits, limitations, integrations, use cases, summary, switching in, who should skip it, our verdict, setup time, pricing fit, features, description
- — changed SSL certificate
- — changed use cases, who it suits, description, features, hidden costs, integrations, limitations, switching in, switching out, models, pricing fit, setup time, who should skip it, summary, platforms, pricing page
- — changed SSL certificate, domain expiry
- — changed who it suits, summary, our verdict, description, integrations, features
- — changed description, setup time, models, our verdict, summary, pricing fit, switching in, who should skip it, switching out, hidden costs, features
Showing the 6 most recent of 12 verification passes.
Free to cite with attribution — this page re-verifies continuously.
How much does N8n Claw cost in 2026?
N8n Claw: Free to use. Pricing last verified 25 Oct 2026.
Where the pricing makes sense
The company stage and team size where N8n Claw's pricing actually pencils out — and where peers do it cheaper.
n8n-claw itself is an open-source, self-hosted project, so your recurring spend lands with your server host and your LLM provider rather than the vendor. For a solo developer on Ollama, that can be near zero marginal cost; for a team routing heavy traffic through Claude or GPT, inference becomes the dominant line. Compare against per-seat managed assistants where cost scales with headcount — the trade is engineering time instead of licences.
Has N8n Claw's pricing changed?
No plan price for N8n Claw has moved since we began logging in Aug 2026; the pricing page is re-read on every verification pass.
Who should pick N8n Claw?
Pick N8n Claw if you are Developers who want a self-hosted AI agent on infrastructure they control, Existing n8n users extending workflows with an autonomous agent, SMEs automating task management and CRM work without per-seat SaaS fees, Privacy-conscious power users avoiding cloud lock-in.
Look elsewhere if you are Non-technical users who want a fully managed SaaS assistant, Teams that need a dedicated mobile app rather than Telegram or API access, Organizations requiring vendor SLAs and enterprise support contracts, Users who want a completely no-code setup with no n8n familiarity.
Skip it if Skip n8n-claw if you want a managed assistant with vendor support and no server to run — it is self-hosted by design and every path to first value starts with your own infrastructure.
GeneratedReal-world workflow fit
Concrete scenarios for the personas N8n Claw actually fits — and what changes day-one when you adopt it.
Clone the repo, run the install script, answer prompts for a Telegram bot token, pick Ollama as the LLM provider, and set a casual personality. Then ask the agent to install the Home Assistant and Notion skills and schedule a daily 8am briefing.
Outcome: A running agent reachable from your phone that nudges you on overdue tasks and controls lights and scenes, with all memory stored on your own disk.
Install the HubSpot, Stripe, and Google Calendar skills by name in chat, then describe recurring actions for pipeline checks and invoice follow-ups. Point the agent at client sites with background checks where monitoring matters.
Outcome: CRM and billing actions happen conversationally, and monitoring runs silently until something actually changes, cutting the noise the team previously got from raw alerting.
Have the agent use its persistent Chromium session to log into a supplier portal, fill a reorder form, and complete the 2FA step by asking you for the code in Telegram. Save the vendor contact into the knowledge graph afterward.
Outcome: A task that used to require a human at a browser runs unattended, and the person and company details are remembered for the next interaction.
Use cases
- Set up a daily 8am briefing and a 15-minute email check purely by describing the schedule in chat.
- Tell the agent to install the HubSpot skill, then create contacts and update deals conversationally.
- Run a silent background monitoring task on a competitor site and only get pinged when the page changes.
- Send a voice note or photo of a meal and have the personal trainer skill log it against a workout plan.
- Have the Chromium browser fill and submit a form or log into a site, relaying the 2FA code through Telegram.
- Delegate a coding task to OpenClaw and receive the finished app back through the agent.
- Join a Google Meet, Teams, or Zoom call via the Vexa Meeting Bot and push live transcription into Notion.
- Keep a persistent markdown project document that survives across conversations and sessions.
Setup time & first value
How long it actually takes to get something useful out of N8n Claw — broken out by persona, not the marketing-page minute.
Solo developer on a prepared server: about 10 minutes from clone to first Telegram message, assuming a bot token and an LLM provider key are ready. n8n user extending existing workflows: roughly 15-30 minutes, since you'll add the agent alongside current automations. Team rolling it out for multiple people: budget a few hours for skill installation, Expert Agent curation, and deciding which
Switching to or from N8n Claw
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From OpenClaw: n8n-claw is positioned as a direct alternative, and it also integrates with OpenClaw if you want to keep using it for autonomous coding tasks.
- →From a managed chat assistant: export the routines you rely on, then recreate them as natural-language recurring actions and reminders in the agent.
- →From zapier-style automations: rebuild the triggers as n8n workflows so the agent can call them as tools, then register them in the skill catalog.
- →From a flat vector-store agent: replace the retrieval layer with the built-in hybrid search and knowledge graph rather than porting embeddings across.
- ↗To a managed assistant: there is no clean path — memory, knowledge graph, and markdown project documents live in your own storage, so plan to export them manually before switching.
- ↗To OpenClaw alone: feasible if your needs are autonomous coding rather than general assistance, since n8n-claw already delegates coding tasks there.
- ↗To a no-code SaaS agent: the skill and Expert Agent definitions have no equivalent format, so expect to recreate integrations from scratch.
Integrations
Resources & guides
Tutorials & Learning
Building an Open Claw Clone in n8n | Full Walkthrough
n8n
n8n-claw: OpenClaw Fully Rebuilt Inside n8n - Autonomous AI Agent with RAG & MCP
Friedemann Schütz
YouTube returned 6 videos for “N8n Claw”, and we withheld 4: 4 did not mention N8n Claw. Showing the 2 we can prove are about N8n Claw.
Official links
What are the best N8n Claw alternatives?
The closest alternatives to N8n Claw we list are Hermes-Agent, GoGogot and xpander.ai. Ranked by direct product-type match, not category overlap.
| Alternative | Pricing | Starts at | Last verified |
|---|---|---|---|
| Hermes-Agent | Freemium | Pricing not published | — |
| GoGogot | Free | Free | — |
| xpander.ai | Freemium | Pricing not published | — |
Alternatives to N8n Claw
View all
Hermes-Agent
Hermes Agent: a self-hosted open-source AI agent with persistent memory that builds reusable skills across 20+ chat platforms.
GoGogot
GoGogot is an open-source, self-hosted AI agent in Go — 8,835 lines, 8 dependencies, 34 tools, answering you in Telegram.
xpander.ai
Enterprise agent runtime that runs Claude Code, Codex, and OpenCode agents inside your own cloud with governance and budgets
Tools that pair well with N8n Claw
Common stack mates teams adopt alongside N8n Claw, with the specific reason each pairing earns its keep.
Hermes-Agent
Hermes Agent: a self-hosted open-source AI agent with persistent memory that builds reusable skills across 20+ chat platforms.
GoGogot
GoGogot is an open-source, self-hosted AI agent in Go — 8,835 lines, 8 dependencies, 34 tools, answering you in Telegram.
xpander.ai
Enterprise agent runtime that runs Claude Code, Codex, and OpenCode agents inside your own cloud with governance and budgets
N8n Claw vs its closest rival
N8n Claw's closest head-to-head is with Locus Robotics — These tools serve entirely different needs. Locus Robotics is a physical warehouse automation system for high-volume fulfillment, costing a significant monthly RaaS fee. N8n Claw is a free, self-hosted AI assistant for developers and SMEs who want an autonomous task manager and chatbot. Choose Locus if you need mobile robots to boost picking productivity; choose N8n Claw if you need a customizable, cloud-free AI agent. Verdict generated
Head-to-head comparisons
N8n Claw vs Locus Robotics
These tools serve entirely different needs. Locus Robotics is a physical warehouse automation system for high-volume fulfillment, costing a significant monthly RaaS fee. N8n Claw is a free, self-hosted AI assistant for developers and SMEs who want an autonomous task manager and chatbot. Choose Locus if you need mobile robots to boost picking productivity; choose N8n Claw if you need a customizable, cloud-free AI agent.
GeneratedN8n Claw vs Presto Voice
If you run a QSR chain and want to boost drive-thru revenue and efficiency, Presto Voice is purpose-built with proven results and a recent Dairy Queen partnership. If you're a developer or team needing an autonomous AI assistant with full data control and zero subscription costs, n8n-claw's self-hosted, free model is a powerful choice. Choose based on your domain and deployment preference.
GeneratedN8n Claw vs Truleo
Choose Truleo if you run a law enforcement agency and need to connect siloed data (RMS, CAD, jail calls, body cameras) into automated case leads. Choose n8n-claw if you want a free, self-hosted AI assistant with adaptive memory and task automation across business tools. They serve completely different domains, so the decision depends entirely on your sector.
GeneratedFrequently asked questions
Used N8n Claw? Help shape our editorial sentiment research.