The Best PulseMCP Alternative for Teams in 2026

PulseMCP is a free MCP server directory that currently lists 3,856 servers, filterable by remote availability, classification (official providers, community, Anthropic references), and popularity. It covers servers from major platforms including Figma, Notion, GitHub, Supabase, Jira, Slack, Confluence, Google, Salesforce, Stripe and Zapier, plus niche offerings across finance, legal, healthcare and more. Each listing shows estimated ecosystem visitors, release dates, popularity rankings, remote endpoint URLs, auth types and transport details, and many include a standardized server.json file with installation instructions. PulseMCP also publishes ecosystem content such as guest posts from MCP server founders on pricing and product strategy. Note that new server submissions and listing changes are currently paused while the site reworks how it ingests and manages listings.

pricing 0

• + Free to browse with no paywall on listings
• + Very large catalog — 3,856 servers spanning coding, business, data and productivity tools
• + Useful filters: remote availability, official-provider classification, popularity and recency sorting
• + Standardized server.json files make installation instructions consistent across listings
• + Estimated visitor and popularity data helps teams gauge which servers are actually used

• − New server submissions and listing changes are paused while ingestion is reworked
• − No security scanning, license checking or benchmarking of listed servers
• − Visitor estimates are algorithmic approximations, not verified usage data
• − It is a discovery directory — teams still need separate tooling to test, secure and monitor what they install

https://www.pulsemcp.com

SkillzBench is AIQ Labs' free, independent directory of AI agent skills plus a free MCP tool that connects any agent to the skills you choose. It is built around five pillars: Discover (continuously crawl and dedupe skills from GitHub), Verify (pin every version to a commit SHA, validate, license-check, security-scan and quality-judge), Benchmark (test every skill in a sandbox with the same model, with and without the skill, publishing the Lift in percentage points), Equip (one free MCP server plus per-agent loadouts where users toggle skills and tools per agent), and Stay current (track upstream changes, re-scan, re-benchmark and notify). Every skill gets a 0-100 trust score built from security, quality, license, maintenance and community signals. The platform addresses published research finding that roughly 26% of skills carry at least one vulnerability. SkillzBench is currently pre-launch and actively being built; the waitlist and newsletter are live.

pricing 0

• + Everything is free — the directory, MCP connector, benchmarks and alerts
• + Security-first: addresses published research finding ~26% of skills carry at least one vulnerability
• + Transparent scoring — Lift is always shown separately from ratings and money never touches scores
• + Per-agent loadouts give teams fine-grained control over what each agent can see and use
• + Continuous monitoring: upstream changes are tracked, with re-scanning, re-benchmarking and notifications

• − Pre-launch — the product is not yet public, so teams cannot use it in production today
• − Focused on agent skills rather than the full breadth of MCP servers PulseMCP catalogs
• − Revenue comes from advertising around blog articles, which some readers may want to weigh even though ads never influence scores

https://skillzbench.com

MCP Hive is a paid subscription platform that bundles access to participating MCP servers through a single plan. The Starter tier costs $19.99/month and includes 8,000 MCP calls — roughly $0.0025 per call — across all participating servers. Professional at $39.99/month raises the allowance to 16,000 calls and adds a $20/month credit for priced MCP servers. Enterprise starts at $249.99/month, includes 16,000 calls, adds capacity in 8,000-call increments for $20/month, includes the $20/month credit, pay-as-you-go beyond credits under a custom spending cap, and enterprise-level customer support. All plans include a 30-day free trial with no credit card required, and subscriptions can be cancelled at any time. On the provider side, MCP Hive retains 30% of what a server earns and pays out the remaining 70%.

pricing 19.99/month · free trial

• + One subscription replaces managing many individual MCP server accounts
• + Transparent per-call economics — about $0.0025 per call at the Starter tier
• + 30-day free trial on all plans with no credit card required
• + Enterprise tier adds pay-as-you-go capacity and a custom spending cap for predictable budgets

• − Monthly call allowances (8,000-16,000) can be tight for agent-heavy workloads
• − Costs scale quickly — Enterprise starts at $249.99/month
• − Access is limited to participating servers, not the entire MCP ecosystem

https://mcp-hive.com

Ref is a premium documentation-search MCP server for AI coding agents, described by its founder as possibly the first standalone MCP server built as a paid product from the ground up. It provides two tools — ref_search_documentation and ref_read_url — each consuming one credit per call, letting coding agents search a public index of library and package documentation and build private indexes of PDFs and internal GitHub repos. Its emphasis is on precise search so agents find exact information without wasting context window. Pricing is credit-based: 200 free credits that never expire, then $9/month for 1,000 credits ($0.009 per search). Since launching this pricing about three months before publication, Ref reported thousands of weekly users and hundreds of subscribers. Its cost structure spans search infrastructure (Turbopuffer), LLM processing in the retrieval pipeline, and web crawling — its highest expense.

pricing 9/month · free trial

• + Search precision tuned for coding agents, reducing wasted context window
• + Private team documentation indexing alongside the curated public index
• + Generous entry point — 200 free credits that never expire
• + Team-friendly: a single shared API key can roll it out across a whole team

• − Single-purpose — documentation search only, not a broad skills or server platform
• − Costs scale with heavy agent usage ($0.009 per search after credits)
• − Positioned as premium versus free alternatives like Context7

https://ref.tools

Context7 is a free MCP server providing basic retrieval-augmented generation (RAG) for documentation, drawing primarily from public GitHub repositories. It is cited in ecosystem pricing discussions as the free baseline against which premium documentation-search products position themselves: Ref's founder describes Context7 as offering basic RAG for documentation from public repos, while Ref sits between Context7 and paid search APIs like Tavily and Exa. For small teams with minimal budget, Context7 represents the zero-cost way to give coding agents some documentation grounding, though its retrieval is positioned as basic rather than precision-optimized, and it does not offer private team documentation indexes.

pricing 0

• + Completely free
• + Simple way to ground coding agents in public documentation
• + No credit system or usage metering to manage

• − Positioned as basic RAG — less precise than premium documentation search tools
• − Primarily limited to public GitHub repo documentation
• − No private indexing of team PDFs or internal repos

Tavily is an agent-focused search and scraping API with a popular MCP server, frequently cited as a pricing benchmark in the MCP ecosystem. Ref's founder names Tavily (alongside Exa) as a key comparable, noting both charge roughly $0.01 per search with volume discounts and both offer popular MCP servers. Tavily serves a different niche than documentation-specific tools: it provides general web search and scraping capabilities that AI agents use to retrieve up-to-date information from across the web, making it relevant for teams building agents that need live web data rather than curated library documentation. Its MCP server is listed among the trending servers in PulseMCP's directory.

pricing See pricing

• + Roughly $0.01 per search with volume discounts
• + Popular, well-established MCP server
• + General web search complements documentation-specific tools

• − Per-search costs add up for high-volume agent workloads
• − General-purpose — not optimized for precise library documentation lookup
• − Pricing details beyond the per-search rate require checking the vendor directly

Exa is an agent-focused search API that, like Tavily, charges roughly $0.01 per search with volume discounts and offers a popular MCP server. It is cited alongside Tavily as one of the two paid search-API benchmarks in the MCP ecosystem, used by Ref's founder to calibrate pricing for documentation-search products. Exa positions itself as search infrastructure for AI agents, providing retrieval capabilities that agents use to pull current information from the web. For small businesses assembling an agent stack, Exa is one of the established names when an agent needs web-scale search through MCP, though per-search pricing means costs scale directly with agent activity.

pricing See pricing

• + Established search provider with a popular MCP server
• + Volume discounts for heavier usage
• + Broad web retrieval for agents

• − Per-search pricing scales with agent activity
• − Not documentation-specific — less precise for library docs lookup
• − Full plan details require checking the vendor directly