Smithery
Smithery is an MCP server marketplace and connection layer that handles OAuth, credentials, and session state for your AI agent.
If you're already building on MCP, Smithery removes the dullest layer of the stack: OAuth handshakes, token refresh, credential storage, and per-server session state, all behind a documented CLI and REST API. The usage-ranked catalog is the genuinely useful signal — Agent News at 43.39k uses and OneSignal at 22.49k tell you what agents actually call, not what markets well. Watch the Arcade.dev ownership for roadmap drift, and don't come here if you're not on MCP yet.
Verified 2h ago · liveness 74/100 · cite: rightaichoice.com/tools/smithery
- Developers building MCP-based agents who want auth, credentials, and sessions handled for them
- Solo founders who need messaging, search, and docs wired up without custom integration work
- AI-native teams using OrgX for task delegation and cross-session memory across agents
- Developers who want to distribute their own MCP servers through a usage-ranked marketplace
- Teams outside the MCP ecosystem with no plans to adopt the protocol
- Organizations that forbid third-party custody of credentials or API keys
- Non-developers who aren't comfortable with MCP setup and agent configuration
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 Smithery if your stack doesn't speak the Model Context Protocol — without MCP clients, the 715-server catalog and managed connection layer have nothing to plug into.
Once you outgrow the free tier, paid-tier pricing isn't itemized on the public pricing page, so you'll only see the real number after creating an account.
The free tier covers the full 715-server marketplace, one-click connections, managed OAuth with token refresh, encrypted credential storage, and server publishing — enough to evaluate whether the connection layer fits your build. Paid tiers exist but aren't itemized on the public pricing page, so you can't compare Smithery's cost against wiring MCP connections yourself or against the individual server vendors' own subscriptions until you create an account. Budget for that unknown before
In short
Smithery — Smithery is an MCP server marketplace and connection layer that handles OAuth, credentials, and session state for your AI agent. Best for Developers building MCP-based agents who want auth, credentials, and sessions handled for them, Solo founders who need messaging, search, and docs wired up without custom integration work, AI-native teams using OrgX for task delegation and cross-session memory across agents. Free to use.
What people actually say about Smithery — 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.
34 mentions across 3 sources (Hacker News, Product Hunt, Lemmy) · researched Aug 18, 2026.
Average across the 3 sources that answered — each source counts once, not each post.
- +One-click connection to 715+ (now 2,000+) MCP servers
- +Managed OAuth with auto token refresh saves hours
- +Encrypted credential storage simplifies security handling
- +CLI and REST API enable programmatic control
- +Centralized session management across agents
- −Path traversal vulnerability exposed 3k servers (Oct 2025)
- −22% of servers flagged vulnerable in community scan
- −Paid tiers hidden behind signup — no upfront pricing
- −Acquisition by Arcade.dev raises long-term uncertainty
- −Server quality varies; ranking doesn't guarantee security
- • No transparent pricing means you must sign up to see costs.
- • Potential extra charges for high-volume usage or multi-agent coordination (speculation).
Viability Score
How well maintained and how widely used is Smithery? 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
- One-click connection to 715 MCP servers from a usage-ranked marketplace
- OAuth, credentials, and session state handled server-side, so your client stays stateless
- Automatic token refresh without manual intervention
- Encrypted write-only storage for API keys and tokens
- Publish your own MCP server for distribution and observability
- CLI for programmatic server connection and management
- REST API for managing MCP connections from your own stack
- Namespaces, organizations, servers, skills, and tokens in the API reference
- OrgX multi-agent coordination: task delegation and cross-session memory
- Flame persistent assistant running on Telegram and Slack
- Per-server usage statistics and popularity ranking across the catalog
- Vercel AI SDK integration
- Agent scaffolds: basic scaffold and ChatGPT App reference implementations
- Works from Cursor, Claude Code, Claude Desktop, and ChatGPT
About Smithery
Smithery is a marketplace and connection layer for Model Context Protocol servers. If your agent needs to send push notifications, search the web, pull version-specific docs, or screen a URL for phishing, you add the server from Smithery's catalog to your toolbox instead of writing OAuth plumbing yourself. The site advertises 715 MCPs and a usage-ranked marketplace, so you can see what other agents actually reach for before you commit. The top of the catalog tells you a lot: Agent News at 43.39k uses, OneSignal at 22.49k, Exa Search at 9.18k, DevMatch at 6.93k, URL Safety Validator at 5.62k, Context7 at 5.56k. Some of these are heavy-lift servers in their own right — DevMatch searches 35M+ developers by verified open-source contributions, DC Hub serves a daily Data Center Power Index across 300+ markets, Niche runs 25 editorial tools for creators. Smithery's job is to make those reachable from Cursor, Claude Code, Claude Desktop, or ChatGPT without you juggling redirect URIs, client IDs, or token refresh. Beyond the catalog, Smithery offers a CLI and REST API for managing connections programmatically, publishing for your own MCP server with distribution and observability, and OrgX for multi-agent task delegation and cross-session memory. Flame is a persistent assistant that runs on Telegram and Slack. Smithery is now part of Arcade.dev, which the site announces in a banner. The honest buyer profile: this is for developers and AI-native teams already committed to MCP. If your stack doesn't speak the protocol, there's nothing here for you — the value is proportional to how many MCP servers you'd otherwise wire up by hand. Compare it against doing each integration directly with the vendor SDK, which gives you more control but none of the shared auth and lifecycle handling.
Behind the Verdict
Where Smithery earns its place is the unglamorous middle. Every MCP server you wire up by hand comes with its own auth story, its own token lifecycle, its own way of breaking when a credential expires at 3am. Smithery puts one connection layer under all of it. For a solo founder shipping a customer-engagement agent that needs push notifications, that's the difference between an afternoon and a week. We'd pick Smithery when you're already committed to MCP and connecting three or more external services. The usage rankings do real work here — when Agent News clears 43k uses and Exa Search sits at 9.18k, you're looking at what other agents reach for, not a curated sponsor slot. Publishing your own server through the same marketplace is a genuine distribution channel, not a checkbox. The case for passing is equally clear. If your stack doesn't speak MCP, Smithery is not the thing that will convince you to adopt it — the value scales with how many servers you'd otherwise maintain. Teams that forbid third-party custody of API keys should stay away, since the whole pitch is that Smithery holds the credentials. And non-developers will bounce off the setup; this is developer tooling with developer ergonomics. The closest alternative is integrating each vendor SDK directly. You get finer control over rate limits and error handling, and you never hand a third party your keys. What you lose is the shared lifecycle — token refresh, stateless connections, per-server usage stats — and you'll rebuild it per integration. For a codebase with two integrations, do it yourself. Past five, the math starts tilting toward Smithery. In practice, the caveat to watch is ownership. Smithery is now part of Arcade.dev, and acquisitions mean roadmap re-prioritization; the catalog and the API are
Researching Smithery? 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 Smithery actually fits — and what changes day-one when you adopt it.
You connect OneSignal through Smithery's REST API and skip writing an OAuth refresh flow, then add Exa Search so the agent can pull current product documentation mid-conversation.
Outcome: Push, email and SMS actions work from a conversation on day one, with Smithery holding the credentials and refreshing tokens for you.
You add Context7 to the toolbox so your assistant fetches version-specific docs and code examples into prompts, and route any link the agent wants to open through URL Safety Validator first.
Outcome: Fewer hallucinated API calls and no agent-follows-malicious-link incidents, with both servers managed under one credential store.
You use OrgX for task delegation, decision approvals and cross-session memory, then publish your own MCP server to the marketplace via the CLI and watch its usage counts.
Outcome: Agents hand off work without losing context, and your server gets distribution plus observability through the marketplace ranking.
Use Cases
- Connect a research agent to Exa Search for live web results, LinkedIn profiles, and deep research.
- Inject version-specific SDK and framework docs into Cursor or Claude Code with Context7 instead of relying on stale training data.
- Screen every link your agent is about to follow with URL Safety Validator for phishing and malware.
- Send push, email, SMS, and in-app messages from OneSignal by describing the audience and message in conversation.
- Query live power-grid, gas, fiber, and ISO telemetry through DC Hub before deciding where to site a data center.
- Source engineering candidates through DevMatch by verified open-source contribution rather than keyword matching.
- Run a persistent assistant in Telegram or Slack via Flame.
- Publish your own MCP server to the marketplace for distribution and usage observability.
Limitations
- Smithery is a connection layer and marketplace for MCP servers, not an AI model — it does not supply the reasoning model your agent runs on.
- Its value is conditional on your stack speaking the Model Context Protocol (MCP); outside that protocol the catalog is inert.
- The public pricing page does not itemize paid tiers, so you need an account before you can see what a paid commitment costs.
- Smithery is now part of Arcade.dev, so roadmap direction is set by the parent company rather than an independent vendor.
as of 2026-10-01
Verification history
We have re-verified Smithery 20 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-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 20 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 Smithery tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free
$0/mo
Ideal for
Developers evaluating MCP or running a small number of agents who want the full catalog before paying anything.
What this tier adds
Starting tier: full access to the 715-server marketplace, one-click connections, managed OAuth with automatic token refresh, encrypted credential storage, session handling, per-server usage stats, and the ability to publish your own MCP server.
Where the pricing makes sense
The company stage and team size where Smithery's pricing actually pencils out — and where peers do it cheaper.
The free tier covers the full 715-server marketplace, one-click connections, managed OAuth with token refresh, encrypted credential storage, and server publishing — enough to evaluate whether the connection layer fits your build. Paid tiers exist but aren't itemized on the public pricing page, so you can't compare Smithery's cost against wiring MCP connections yourself or against the individual server vendors' own subscriptions until you create an account. Budget for that unknown before
Setup time & first value
How long it actually takes to get something useful out of Smithery — broken out by persona, not the marketing-page minute.
Developers already using an MCP client: browse the catalog, click Add to toolbox, and you're connected in minutes — zero OAuth configuration means no redirect URI or client-secret dance. Teams wiring connections into their own app via the REST API should budget an afternoon against the API reference (servers, skills, tokens, namespaces, organizations). Publishing your own MCP server takes longer,
Switching to or from Smithery
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From hand-rolled OAuth integrations: move each service into Smithery's toolbox and let managed token refresh replace your credential code.
- →From a self-maintained MCP server list: point your client at Smithery's catalog and use the usage ranking to prioritize which servers to connect.
- →From direct Exa or Context7 API keys in your own code: connect the equivalent MCP server and drop the keys from your environment.
- →From a chat-tab-only assistant: move it to a persistent Telegram or Slack presence via Flame.
- ↗To direct vendor SDKs: if you outgrow managed credentials, reimplement auth against each service's own API and remove Smithery from the connection path.
- ↗To self-hosted MCP servers: if third-party credential custody becomes a blocker, run the servers yourself and manage tokens in your own secrets store.
- ↗To Arcade.dev's other products: as Smithery's parent, Arcade.dev may offer adjacent integration paths worth evaluating alongside it.
Integrations
Resources & Guides
- Documentationsmithery.ai
Introduction
Smithery gives agents more agency by connecting them to the outside world — search the web, query databases, control browsers, reach into business tools, and much more.
- Documentationsmithery.ai
Llms
Full product docs from smithery.ai
- Documentationsmithery.ai
Mcp
Full product docs from smithery.ai
- Documentationsmithery.ai
Index
Full product docs from smithery.ai
Tutorials & Learning
YouTube returned 6 videos for “Smithery”, and we withheld 6: 6 could not be judged, because “Smithery” 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 Smithery.
Official links
Tools that pair well with Smithery
Common stack mates teams adopt alongside Smithery, with the specific reason each pairing earns its keep.
Albato
Embedded iPaaS that white-labels 1,000+ native integrations, customer-facing AI agents, and a Universal MCP inside your SaaS.
Nubase
Open-source, self-hosted backend that lets AI coding agents ship full-stack apps from chat
RoboWork
One interface for ChatGPT, Claude, Gemini, and Grok with agent building, prompt library, and MCP plugin support.
Featured Head-to-Head Comparisons
Backengine Mcp vs Smithery
If your bottleneck is private company knowledge—support docs, wikis, procedure manuals—BackEngine MCP is the focused choice: it turns that internal data into AI-queryable assets with security boundaries. But if you're building agents that need a wide variety of external tools and you want to skip auth headaches, Smithery's 715+ server marketplace and managed OAuth make it the pragmatic pick. Choose based on whether your data is internal (BackEngine) or your needs are external (Smithery).
Ekos vs Smithery
If you want to supercharge your existing GitHub repos with AI, EKOS is the direct route. If you're building agents that need a universe of pre-built tools without auth headaches, Smithery is unbeatable. Choose based on whether you're repo-centric or agent-centric.
Fetchsandbox Mcp vs Smithery
If your pain is trusting AI-generated integration code, FetchSandbox MCP is the focused, safety-first choice — it proves fixes work in isolation before they touch production. If you're building agents and need fast access to a broad tool ecosystem with auth handled for you, Smithery is the pragmatic pick. Pick FetchSandbox for validation rigor, Smithery for breadth and speed.
Alternatives to Smithery
View all
Frequently Asked Questions
Categories
Best-of guides
Used Smithery? Help shape our editorial sentiment research.