PulseMCP vs SkillzBench: Which Agent Skills Platform Should You Trust?
PulseMCP is a large, free-to-browse directory of MCP servers — its catalog lists 3,856 servers with filters for official providers, remote availability, and popularity, though new submissions were paused while its listing pipeline is reworked. SkillzBench is a free, independent directory of AI agent skills (SKILL.md packages) that verifies, security-scans, and benchmarks every skill, publishing a measured Lift percentage and a 0–100 trust score, plus a free MCP connector to load chosen skills into your agents. If you need to find an MCP server for a specific tool, PulseMCP's breadth wins; if you need to know whether a skill is safe and actually improves your agent, SkillzBench's verification and benchmarking are the differentiator.
TL;DR
- PulseMCP is a free MCP server directory listing 3,856 servers, filterable by official providers, remote availability, and popularity
- New server submissions and listing changes on PulseMCP are paused while the ingestion pipeline is reworked
- SkillzBench is a free directory of AI agent skills with verification, security scanning, and sandbox benchmarking that publishes a Lift score in percentage points
- SkillzBench pins every skill version to a commit SHA and runs license checks, quality judging, and security scans before listing
- SkillzBench's trust score (0–100) is built from security, quality, license, maintenance, and community signals — money never touches scores
- SkillzBench is pre-launch: the waitlist is live, and the directory, MCP connector, benchmarks, and alerts will all be free
- PulseMCP suits server discovery; SkillzBench suits skill vetting — many teams will use both
Head to head
| Dimension | PulseMCP | SkillzBench |
|---|---|---|
| Pricing | Free to browse the directory; individual servers list their own costs | Free — directory, MCP connector, benchmarks, and alerts all free, ad-funded |
| Core purpose | Discovery of MCP servers for MCP clients | Verification, benchmarking, and equipping of agent skills |
| Catalog scope | 3,856 MCP servers with filters and popularity rankings 🏆 | Agent skills (SKILL.md) crawled from GitHub, deduplicated continuously |
| Trust & verification | Official vs community classification; no published security scanning | Security scans, license checks, commit-SHA pinning, 0–100 trust score 🏆 |
| Benchmarking | None published | Sandbox tests with/without each skill, publishing Lift in percentage points 🏆 |
| Freshness & maintenance | Historically updated daily; submissions and listing changes currently paused | Tracks upstream changes, re-scans, re-benchmarks, and notifies users 🏆 |
| Current availability | Live and browsable now 🏆 | Pre-launch; waitlist and newsletter live, product still being built |
The contenders
PulseMCP operates one of the largest public directories of Model Context Protocol (MCP) servers, built to work with a variety of MCP clients. The catalog lists 3,856 servers and is filterable by classification (official providers vs community), remote availability, last-updated date, and popularity rankings across all-time, monthly, and weekly windows. Each server page carries practical detail: provider, classification, estimated visitor/download counts, release date, popularity rank, remote endpoint URLs with auth requirements, transport type (including the current Streamable HTTP spec), and cost indicators — some servers are explicitly marked entirely free with no upsells. A standardized server.json file is available for many listings, giving install instructions and configuration options in a copy-paste-friendly format. The site also surfaces trending categories (Figma, Notion, GitHub, Slack, Stripe, Shopify, Google Calendar, and dozens more) and publishes editorial content, including guest posts from MCP builders on pricing and product strategy. One current caveat: new server submissions and listing changes are paused while PulseMCP reworks how it ingests and manages listings.
pricing See pricing
- + Free to browse with no account or subscription required for the directory
- + Very large catalog — 3,856 servers covering mainstream tools like Slack, Stripe, Shopify, Notion, and Google Calendar
- + Rich per-server detail: endpoints, auth type, transport, cost, and server.json install files
- + Popularity and visitor estimates help gauge which servers are actually used
- + Official vs community classification helps identify vendor-maintained servers
- − New server submissions and listing changes are currently paused during a rework of the ingestion pipeline
- − No published security scanning, license checking, or benchmarking of listed servers
- − Visitor and popularity figures are algorithmic estimates, not audited usage data
- − Discovery-focused: it tells you a server exists but not how well it performs against alternatives
SkillzBench, built by AIQ Labs, is a free, independent directory of AI agent skills — SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents — plus a free MCP tool that connects any agent to the skills you choose. Its premise is that skills are proliferating faster than anyone can vet them: published research found roughly 26% of skills carry at least one vulnerability. SkillzBench is built around five pillars. Discover continuously crawls and dedupes skills from GitHub. Verify pins every version to a commit SHA, validates it, checks the license, runs a security scan, and applies a quality judgment. Benchmark tests every skill in a sandbox with the same model, with and without the skill, and publishes the Lift in percentage points. Equip provides one free MCP server plus per-agent loadouts — you toggle skills and tools per agent, and the agent only sees what you chose. Stay current tracks upstream changes, re-scans and re-benchmarks, and notifies you. A 0–100 trust score combines security, quality, license, maintenance, and community signals, with Lift always shown separately and money never touching scores. The platform is pre-launch: the waitlist is live, and everything — directory, MCP connector, benchmarks, alerts — will be free, funded by advertising around blog articles that never influences scores or listings.
pricing 0
- + Everything is free — directory, MCP connector, benchmarks, and alerts
- + Directly addresses the ~26% of agent skills found to carry at least one vulnerability
- + Lift benchmarking shows measurable performance impact, not just marketing claims
- + Commit-SHA version pinning means you know exactly which code you're running
- + Money never touches scores or listings — ad-funded, not pay-to-rank
- + Per-agent loadouts give precise control over what each agent can access
- − Pre-launch: the product is still being built, so the directory and MCP tool are not yet publicly available
- − Focuses on agent skills (SKILL.md packages) rather than the broader MCP server ecosystem
- − No track record of published benchmark results yet, since the platform hasn't launched
- − Waitlist is currently the only way to get early access
Verdict
Winner: none
These platforms solve different halves of the same problem: PulseMCP helps you discover MCP servers at scale, while SkillzBench focuses on verifying and benchmarking agent skills so you know which ones are safe and effective. Neither is strictly better — the right pick depends on whether you're shopping for servers or vetting skills.
| Option | Overall | Breakdown |
|---|---|---|
| PulseMCP | 7.5 | : 9: 8: 7: 5: 6: 8 |
| SkillzBench | 8 | : 10: 7: 9: 10: 8: 8 |
What each platform actually does
Trust and verification compared
Cost and business model
Which should you pick?
- :
- :
- :
- :
- :
FAQ
Is PulseMCP free to use?
Yes — browsing the PulseMCP directory is free. The site lists 3,856 MCP servers with filters, popularity rankings, and per-server detail pages including endpoints, auth types, and cost indicators. Individual servers listed on PulseMCP set their own costs, and many are marked entirely free, so your overall spend depends on which servers you adopt, not on PulseMCP itself.
How much does SkillzBench cost?
Everything on SkillzBench is free: the skills directory, the MCP connector, sandbox benchmarks, and upstream-change alerts. The platform is funded by advertising shown in and around its blog articles, and that ad revenue never influences trust scores or listings. Note that SkillzBench is pre-launch — the waitlist and newsletter are live while the product is being built.
Can I use PulseMCP and SkillzBench together?
Yes, and for many teams that's the natural setup. PulseMCP helps you discover MCP servers to connect agents to external tools, while SkillzBench helps you vet and equip agent skills — the SKILL.md instruction packages for Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents. One covers server discovery; the other covers skill verification, benchmarking, and controlled deployment via per-agent loadouts.
Why does skill verification matter for small businesses?
Published research found that roughly 26% of agent skills carry at least one vulnerability. For a small business adopting AI agents without a dedicated security team, installing an unvetted skill can expose data or degrade agent behavior. SkillzBench addresses this with commit-SHA version pinning, security scans, license checks, quality judging, a 0–100 trust score, and sandbox benchmarks that publish each skill's Lift in percentage points.
Is SkillzBench available now?
Not yet publicly. SkillzBench is pre-launch and actively being built — the waitlist and newsletter are live, but the directory, MCP connector, benchmark results, and alerts are what the platform is building and will offer. Everything is planned to be free, funded by blog advertising that never touches scores or listings. PulseMCP, by contrast, is live and browsable today, though its new server submissions are currently paused during a rework.