Field-by-field comparison

Compare the evidence before choosing an MCP.

Each signal stays separate, missing facts remain visible, and the column order follows your selection.

MCPs in this comparison

Uncheck an MCP to remove it. Comparing two or three keeps each signal in its own row.

Open canonical comparison link

Comparable alternatives

Shown only when capability or industry context overlaps.

Side-by-side MCP evidence comparison
Evidence fieldFirecrawl MCPMCP100 RecommendedHubSpot MCPMCP100 Recommended
Task fit
Best for
  • Scraping and extracting public web pages
  • Search, crawl, map, batch, and research workflows
  • Teams already using Firecrawl API limits and governance
  • Querying CRM records, activities, marketing content, and organizational context
  • Creating or updating supported CRM records and activities with confirmation
  • Building managed AI workflows around HubSpot account data
Not ideal for
  • Simple known-page fetches that do not justify a paid API
  • Sites whose terms or access controls prohibit automated collection
  • Workflows requiring guaranteed freshness or extraction accuracy
  • Clients that do not support OAuth 2.1 with PKCE
  • Vector or semantic CRM search, which the documented backend does not provide
Avoid when
  • You cannot govern which URLs the agent may access
  • The workflow may collect personal, confidential, copyrighted, or access-controlled content without review
  • Unexpected API-credit consumption is unacceptable
  • Existing HubSpot user permissions are broader than the intended agent workflow
  • CRM or communication records cannot be shared with the chosen AI client
  • Record creation, email logging, or meeting actions cannot require human approval
Provenance
Provenance details

First-party MCP

Firecrawl · Publisher source ↗

First-party MCP

HubSpot · Publisher source ↗

Maintenance
Maintenance details

Repo: Sep 24, 2026

Package: Sep 21, 2026

Repo: Not documented

Package: Not documented

Popularity evidence
GitHub stars

7,507

GitHub stars · checked 2026-09-24T04:14:47.690000Z

Not documented
Package downloads / 30 days

126,701

package downloads / 30 days · checked 2026-09-24T04:17:59.638000Z

No reliable download data
External adoption evidence
  • npm: package downloads: 412,488Exact attributable signal.npm: package downloads · Rolling 30 days · checked 2026-08-14T18:46:26.000ZEvidence source ↗
Not documented
Client coverage
Client coverage details
Claude Code✓ Local stdio
Claude Desktop✓ Local stdio
Codex✓ Local stdio
VS Code✓ Local stdio
Cursor✓ Local stdio
OpenCode✓ Local stdio
Claude Code✓ Streamable HTTP
Claude Desktop✓ Streamable HTTP
Codex✓ Streamable HTTP
VS Code✓ Streamable HTTP
Cursor✓ Streamable HTTP
OpenCode✓ Streamable HTTP
Client coverage noteLocal stdio runs on your computer. Streamable HTTP connects to a remotely hosted MCP server.
Access and data
AuthenticationCloud and local-package configurations use FIRECRAWL_API_KEY. Official docs also show hosted MCP URLs containing the key; environment-variable or protected input handling is safer than embedding the key in a URL.OAuth 2.1 with PKCE is required. A HubSpot MCP auth app supplies client ID, client secret, and registered redirect URL; the user selects an account and grants available permissions.
CostMCP calls consume the account's standard Firecrawl API credits and rate limits. Search, crawl, batch, and agent operations can consume more resources than single-page scraping; current plan pricing applies.HubSpot documents the remote MCP server as generally available to all HubSpot accounts without a separate MCP fee. HubSpot subscription features, API limits, integration hosting, and AI-client costs still apply.
PermissionsThe MCP can initiate outbound requests to user- or model-selected URLs and invoke Firecrawl scrape, search, crawl, map, extraction, batch, and research capabilities according to enabled tools and account limits.Read and write availability is governed by the MCP app's granted scopes together with the connected user's HubSpot permissions. Supported actions can include CRM records, activities, marketing content, and selected mutations; do not describe access as an over-specific universal intersection beyond those app scopes and user permissions.
Data handlingURLs, queries, extraction schemas, and retrieved page content are processed by the configured Firecrawl service. Cloud mode sends them to Firecrawl; self-hosted mode follows the operator's deployment and any configured model/provider dependencies.HubSpot's hosted MCP service relays authorized CRM and activity data to the MCP client. Sensitive Data accounts block activity objects through MCP. Client credentials, OAuth tokens, retrieved records, and tool results must be protected by the integrating application.
Limitations
Tradeoffs
  • Broad scrape, crawl, search, and autonomous-agent tools are powerful but enlarge prompt-injection, compliance, and cost exposure.
  • Cloud mode is easy to configure; self-hosting adds operational burden.
  • Remote URL forms that embed an API key are convenient but can leak through configuration, logs, screenshots, or history.
  • The reviewed v2.0.0 release moved the MCP server to the Firecrawl v2 API and added Streamable HTTP.
  • The GA hosted service removes local server maintenance but requires creation and management of an MCP auth app.
  • Available scopes follow current server tools and user-granted permissions; newly added scopes can require reauthorization.
  • HubSpot blocks activity objects through MCP when Sensitive Data is enabled, reducing exposure but also capability.
Risk contextMain risks are prompt injection from untrusted pages, collection-policy violations, sensitive URL/query disclosure, API-key leakage, and agent-driven credit spend. Use URL allow-lists, low limits, human approval, and a dedicated low-budget key.CRM and activity tools can expose personal and commercially sensitive data or create consequential customer records and communications. Use a dedicated auth app, narrow user permissions, secure PKCE and token storage, re-review scope changes, and confirm mutations. No MCP100 execution testing is claimed.
Evidence date
Editorial review2026-09-052026-07-24
Candidate evidence2026-09-24T04:25:05.554391Z2026-09-24T04:25:05.554391Z

Popularity, maintenance, fit, permissions, and client support are independent evidence fields.

Read the evidence method →