Question 1

How does SkillzBench measure whether an agent skill actually improves performance?

Accepted Answer

SkillzBench's Benchmark pillar tests every skill in a sandbox using the same model, running the task with and without the skill, then publishes the Lift in percentage points. That Lift is always shown separately from ratings and Trust scores, so performance data can never be inflated by community sentiment or paid placement — money never touches scores.

Question 2

Can SkillzBench help with MCP tool poisoning and supply-chain risks before we deploy skills to client agents?

Accepted Answer

The Verify pillar pins every skill version to a commit SHA, then runs validation, license checks, security scans, and quality judgments. A 0-100 Trust score aggregates security, quality, license, maintenance, and community signals. Note that at launch, scope covers instruction-only skills (SKILL.md plus text references); script-bearing skills are listed link-out only.

Question 3

How does SkillzBench handle the fact that most marketplace skills are abandoned or invalid?

Accepted Answer

The Discover pillar continuously crawls and dedupes skills from GitHub, so you're not sifting through placeholder projects. Maintenance and community signals feed into the Trust score, and the Stay current pillar tracks upstream changes — if a repo is abandoned or changes underneath you, SkillzBench re-scans, re-benchmarks, and notifies you.

Question 4

When can my agency actually start using SkillzBench, and what will pricing look like?

Accepted Answer

SkillzBench is pre-launch and actively being built — the waitlist/newsletter is live today, and joining it is the fastest path to early access. The pricing hypothesis includes a generous free tier, Pro at roughly $9-15/mo for more agents and unlimited enabled skills, Teams at roughly $15-25/seat/mo with org loadouts, allowlists, and audit, and a Creator tier for verified authors.

Question 5

How is SkillzBench different from just picking skills off the MCP Registry or a marketplace?

Accepted Answer

Registries and marketplaces list skills; SkillzBench tests them. You get commit SHA pinning, security scans, license checks, quality judgments, a 0-100 Trust score, and sandboxed Lift benchmarks — plus one MCP server with per-agent loadouts so each agent only sees the skills and tools you chose. It's an independent lab and control panel, not a listing site with a sales agenda.

MCP Server Discovery for Agencies