# Fuse MCP (slim)

> Fuse MCP is the hosted MCP server for Fuse AI's sales platform. It lets an assistant search people and companies, enrich contacts, manage lists and campaigns, and send email and LinkedIn messages from a signed-in Fuse account.

- Full: https://www.anchorterminal.com/tools/fuse-mcp.md (~7,400 tokens) · this version ~1,830 tokens · JSON https://www.anchorterminal.com/tools/fuse-mcp.json · canonical https://www.anchorterminal.com/tools/fuse-mcp
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-08

**D · 47.9/100 · rank #543 of 629 · #11 in Lead & company data · not agent-ready · confidence medium**


Assessment: The server documents 290 tools with read-only and destructive annotations, OAuth with PKCE and dynamic client registration, and published limits of 60 requests a minute. The single `mcp:tools` scope grants everything the user can do, sends happen at once with no undo, and no status page, changelog or security contact was found.

## Facts

- Kind: MCP server · vendor: FuseAI, LLC · category: Lead & company data · legal entity: FuseAI, LLC · provenance 62/100
- Endpoint: `https://mcp.fuseai.com` (Streamable HTTP)
- Auth: OAuth · pricing: Freemium · x402: no · licence: Proprietary hosted service under FuseAI's terms of service. No public source repository found
- Probe metrics: not measured yet (probes haven't run)
- Surface graded: The hosted MCP server at https://mcp.fuseai.com. The REST API at https://api.tryfuse.ai/api/v1/business is a separate surface with its own keys and limits
- Tools: 290 as listed on 7 October 2026. Prospect search and enrichment, lists, campaigns, inbox, email accounts and sending domains, LinkedIn, signal agents, Smart Actions and tasks, website visitors, power dialer, automations, CRM sync and Slack, research, workspace and team
- Credentials: OAuth authorisation code with PKCE (S256) for public clients, dynamic client registration at https://core.backend.fuseai.com/oauth/register, single scope `mcp:tools`. Access tokens last one hour, refresh tokens rotate and last 30 days from last use
- Rate limits: 60 requests a minute and 5,000 a day per user, counting the handshake and tool listing. A batch counts once. 429 with `RATE_LIMIT_EXCEEDED`
- Credits: Search 2 credits a profile or company, up to 50 a work email found (100 when personal emails are prioritised), 200 a phone, 10 a re-verified email, Deep Research capped at 1,000, 2,000 or 5,000 a run. Reads are free. Every call needs a balance of at least 100 credits
- Errors: `isError: true` with a JSON `errors` array of `code` and optional `message`, such as `TOOL_ARGUMENT_VALIDATION_FAILED` or `LIST_NOT_FOUND`. Four plain-text messages cover unknown tools, low credits, billing and timeouts
- Timeouts: 60 seconds a tool call. Imports, enrichment, smart columns and Deep Research answer at once and finish in the background
- Annotations: `readOnlyHint` on tools that only read and `destructiveHint` on tools that delete, remove, stop, disconnect or overwrite. Sends, launches and CRM pushes are not marked destructive
- Context cost: About half a megabyte of tool descriptions and input schemas per the vendor. No toolsets or read-only subset
- Data providers: 39 named in the docs for Deep Research, among them Apollo, People Data Labs, Hunter, Lusha, Prospeo, FullEnrich and ZeroBounce. The site says 40+
- CRM sync: Attio, HubSpot, Salesforce, Zoho and Pipedrive, plus Slack messages
- REST API: OpenAPI 3.1, version 1.0.0, 118 paths and 146 operations. HTTP Basic with an `af_` key, nine scopes fixed at creation, optional expiry, limits per surface from 20 to 120 a minute
- Plans: Free $0, Solo $60 a month (480,000 credits a year, 1 seat), Team $200 a month (1,200,000 credits a year, 5 seats), Enterprise by quote with RBAC, SSO and audit logging
- Prices: Free plan free per month (plan); Solo plan $60 per month (plan); Team plan $200 per month (plan)

- Scores: Reliability 40, Performance pending, Schema & documentation 72, Agent ergonomics 57, Security & auth 45, Payments & pricing 30, Task success pending, Maintenance & community 35, Transparency & trust 49 · total over the 7 assessed categories
- Why: Reliability, Graded as a hosted service. · Schema & documentation, The docs say every tool ships a description and an input schema, about half a megabyte in all. · Agent ergonomics, 290 tools and about half a megabyte of definitions, with no toolsets, dynamic loading or read-only subset. · Security & auth, OAuth authorisation code with PKCE and dynamic client registration, one-hour access tokens and rotating refresh tokens. · Payments & pricing, No x402, MPP or L402 (0). · Maintenance & community, The tool reference is dated 7 October 2026 and the tool list was rebuilt in October 2026 (30). · Transparency & trust, Closed service with published terms of service, last updated 1 July 2026, naming FuseAI, LLC (15).
- Sources: 22, open questions: 8, both in the full twin
- Capabilities: lead.search, lead.enrichment, email.finder, email.verification, data.company, email.send, search.research
- JSON: https://www.anchorterminal.com/api/v1/tools/fuse-mcp.json
- Verify (for the vendor): the badge `https://www.anchorterminal.com/badges/fuse-mcp.svg` or a link to https://www.anchorterminal.com/tools/fuse-mcp from a page on fuseai.com or one of its subdomains, then `POST https://www.anchorterminal.com/api/v1/verify` `{"slug", "url"}` or `verify_listing` at /mcp; re-checked weekly, no effect on the grade. Snippets in the full twin.

## Before you call it

1. Keep the client's approval prompt on for every tool that is not read-only. Sends, CRM pushes and Slack posts run at once with no undo
2. Launch a campaign only on an explicit request. Creating one makes a draft, and `update_campaign_status_tool` starts real outreach
3. Call `get_credits_balance` and `get_credit_prices` first. Every call needs at least 100 credits, and searches cost 2 credits a result
4. Resolve filter values with `lookup_filter_values` or `get_company_search_enums` before `search_people` or `search_companies`
5. Start long work with `deep_research_start`, then poll `deep_research_status`. A tool call has 60 seconds, and runs take 5 to 20 minutes

## Connect

```bash
claude mcp add --transport http fuse https://mcp.fuseai.com
```
Full config and headless snippets are in the full page. Through letme (picks today, calling later): https://letme.dev/fuse-mcp

## Similar tools

| Tool | Grade | Score | Shared capabilities | Slim |
| --- | --- | --- | --- | --- |
| LeadMagic API + MCP | C | 60.2 | lead.search, lead.enrichment, email.verification, email.finder, data.company | https://www.anchorterminal.com/tools/leadmagic.min.md |
| Hunter API + MCP | C | 56.5 | email.finder, email.verification, lead.search, lead.enrichment, data.company | https://www.anchorterminal.com/tools/hunter.min.md |
| Manifold | C | 56.2 | lead.search, lead.enrichment, email.finder, email.verification, data.company | https://www.anchorterminal.com/tools/manifold-mcp.min.md |
| Apollo API + MCP | B | 63.3 | lead.search, lead.enrichment, email.finder, data.company | https://www.anchorterminal.com/tools/apollo.min.md |
| Coresignal API + MCP | B | 62.9 | lead.search, lead.enrichment, email.finder, data.company | https://www.anchorterminal.com/tools/coresignal.min.md |


## Panel reviews (0, desk reviews from public material, no calls made)