Is Agenticskills.io Worth It in 2026? Security, Licensing and Maintenance Compared
Agenticskills.io is a free, editorially curated directory indexing 193+ agent skills and 196 MCP servers across 9 platforms, with an 11-guide learning library and a stated policy that listings are never paid for — a solid resource for discovering skills, though its rankings are editorial judgment rather than measured benchmarks. SkillzBench, currently in pre-launch, is building a free alternative that pins every skill version to a commit SHA, security-scans and license-checks each one, and benchmarks skills in a sandbox with and without the skill to publish a measured Lift. For small businesses that need evidence a skill actually helps and isn't risky, SkillzBench's verify-and-benchmark approach is the one to watch.
TL;DR
- Agenticskills.io is live now with 193+ curated skills, 196 MCP servers and 9 platforms tracked (index v1.7.0, Oct 13 2026).
- Agenticskills.io states listings are never paid for and that its rankings are editorial judgment, not measurement.
- Its learning library contains 11 guides totaling 177 minutes and 41K+ words, including a security analysis of 23,000+ public MCP servers where 73% failed basic security checks.
- SkillzBench is pre-launch; its waitlist/newsletter is live but the product is not yet public.
- SkillzBench is building commit-SHA version pinning, license checks, security scans and sandbox benchmarking that measures each skill's Lift with and without the skill.
- SkillzBench cites published research that ~26% of skills carry at least one vulnerability — the core problem it is designed to solve.
- Both are free: Agenticskills.io is free to browse, and SkillzBench plans everything free, funded by blog advertising that never influences scores.
Head to head
| Dimension | SkillzBench | Agenticskills.io |
|---|---|---|
| Pricing | Everything free (directory, MCP, benchmarks, alerts); ad-funded, ads never touch scores | Free to browse; free-forever newsletter |
| Security | Building security scanning into the verification pipeline for every skill; motivated by research that ~26% of skills carry a vulnerability 🏆 | Publishes security research (73% of 23,000+ audited MCP servers failed basic checks) and a 12-step audit framework; no stated per-skill scanning pipeline |
| Licensing | License-check is an explicit step in the Verify pipeline, plus a license signal in the trust score 🏆 | Shows license per tool entry (e.g. Fly.io: Proprietary) |
| Maintenance | Plans to track upstream changes, re-scan, re-benchmark and notify 🏆 | Live versioned index (v1.7.0, Oct 13 2026); no stated upstream change tracking |
| Benchmarking | Sandbox benchmarking with the same model, with and without the skill; Lift published in percentage points 🏆 | Rankings are editorial judgment, not measurement (stated on-site) |
| Content & education | Blog in development alongside the platform | 11 guides, 177 minutes, 41K+ words, plus workflow guides and a marketplace guide 🏆 |
| Availability | Pre-launch; waitlist/newsletter live | Live now 🏆 |
| Transparency | Lift shown separately from ratings; money never touches scores | States listings are never paid for and labels rankings as editorial judgment |
The contenders
Agenticskills.io is a curated directory of AI agent skills and MCP servers, live as of index version 1.7.0 (October 13, 2026) with 193+ skills indexed, 196 MCP servers and 9 platforms tracked. It covers skills for Claude Code, OpenAI Codex, Cursor, Gemini CLI and other agents that read SKILL.md files. The site organizes skills into 16 categories with editor's picks, S-rank and A-rank designations, and install counts (featured entries range from 4.0K to 180.7K). Alongside the directory it publishes an in-depth learning library — 11 guides totaling 177 minutes and 41K+ words — covering the npx skills CLI, Claude Skills, the Agent Plugins open standard, and security topics including an analysis of 8 independent audits covering 23,000+ public MCP servers in which 73% failed basic security checks. The site states that listing is never paid for and that its rankings are editorial judgment rather than measurement.
pricing 0
- + Live today with a large index — 193+ skills, 196 MCP servers, 9 platforms
- + Explicit, stated policy that listings are never paid for
- + Deep editorial content, including a 12-step framework for auditing MCP servers before install
- + Transparent methodology disclosure: rankings are labeled editorial judgment, not measurement
- + Covers the full ecosystem including the Agent Plugins open standard backed by OpenAI, Microsoft, AWS, GitHub, Cursor and Vercel
- + Free to browse, with a free-forever newsletter
- − Rankings are editorial judgment, not measured benchmarks — the site says so itself
- − No stated per-skill security scanning, license verification or version pinning pipeline
- − No sandbox benchmarking that measures whether a skill actually improves agent output
- − No stated mechanism for tracking upstream skill changes or notifying users of updates
- − No trust score combining security, license and maintenance signals
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 currently pre-launch — the waitlist and newsletter are live, and the product is actively being built. Its pipeline 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, and publish 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). A trust score from 0 to 100 combines security, quality, license, maintenance and community signals; Lift is always shown separately from ratings and money never touches scores. The motivation is published research finding that roughly 26% of skills carry at least one vulnerability. Everything on the platform — directory, MCP connector, benchmarks and alerts — will be free, funded by advertising shown in and around blog articles that never influences scores or listings.
pricing 0
- + Verification pipeline designed around the exact risks the title asks about: security, licensing and maintenance
- + Measured benchmarks — every skill tested in a sandbox with and without the skill, with Lift published in percentage points
- + Commit-SHA pinning means you always know exactly which version was scanned and benchmarked
- + Upstream change tracking with re-scanning and notifications keeps assessments current
- + Everything free — directory, MCP connector, benchmarks and alerts — with ad money never influencing scores
- + Per-agent loadouts let an agent see only the skills and tools explicitly chosen for it
- − Pre-launch: the product is not yet public, so none of its features can be used today
- − No live skill counts, user counts or published benchmark results yet
- − Only the waitlist/newsletter is currently available
Verdict
Winner: Agenticskills.io
Agenticskills.io is live today with a large indexed directory and deep editorial content; SkillzBench is pre-launch and cannot yet be used. On the specific dimensions the title asks about — security, licensing and maintenance — SkillzBench's planned commit-SHA pinning, license checks and re-scanning is the stronger model, but it is not yet public.
| Option | Overall | Breakdown |
|---|---|---|
| Agenticskills.io | 7.5 | : 9: 6: 6: 6: 9 |
| SkillzBench | 7 | : 10: 9: 9: 9: 2 |
Security compared
Licensing and maintenance compared
Content, discovery and cost
Which should you pick?
- :
- :
- :
- :
- :
FAQ
Is Agenticskills.io free to use?
Yes. The directory of 193+ skills and 196 MCP servers is free to browse, and its newsletter is described as free forever with unsubscribe anytime. The site also states that listings are never paid for, so inclusion isn't influenced by payment.
Does Agenticskills.io security-test the skills it lists?
Not systematically, as far as its own pages state. It publishes strong security education — an analysis of 8 audits covering 23,000+ MCP servers where 73% failed basic checks, plus a 12-step audit framework — but its tool pages explicitly describe inclusion as editorial judgment, not measurement. Per-skill scanning is the gap SkillzBench is being built to fill.
What is SkillzBench and when does it launch?
SkillzBench is AIQ Labs' free, independent directory of AI agent skills plus a free MCP connector. It is currently pre-launch: the waitlist and newsletter are live while the platform is actively built. It plans commit-SHA version pinning, license checks, security scans, sandbox benchmarking with published Lift, and upstream change tracking with notifications.
How does SkillzBench's benchmarking work?
Every skill is tested in a sandbox with the same model, twice: once with the skill and once without. The difference is published as a Lift in percentage points, always shown separately from ratings, so you can see exactly how much a skill measurably helps rather than relying on opinion scores.
Why does skill security matter for a small business?
Because agent skills are instruction packages your AI agent executes, and published research found roughly 26% of skills carry at least one vulnerability. A compromised or poorly maintained skill can leak credentials or break workflows — risks a small business without a security team is poorly placed to catch manually.
Can I use both Agenticskills.io and SkillzBench together?
Yes, and that's a sensible approach. Use Agenticskills.io today for discovery and its 11-guide learning library, then run candidates through SkillzBench's verification and benchmarking once it launches — its free MCP connector is designed to equip agents with only the skills you've chosen and vetted.