Question 1

What does SkillzBench actually verify in a SKILL.md agent skill?

Accepted Answer

SkillzBench is building a Verify pillar that pins every skill version to a commit SHA, then validates the package, checks its license, runs security scans, and applies a quality judgment. Launch scope covers instruction-only skills (SKILL.md plus text references); script-bearing skills will be listed link-out only. Results feed into a 0-100 trust score built from security, quality, license, maintenance and community signals.

Question 2

How is the Lift benchmark calculated for agent skills?

Accepted Answer

Each skill is tested in a sandbox using the same model, run with and without the skill, and SkillzBench publishes the difference as a Lift in percentage points. Lift is always shown separately from trust ratings, so a well-marketed skill can't hide behind a good score — and money never touches either number.

Question 3

Can I control which skills and tools each agent can see?

Accepted Answer

Yes — the Equip pillar provides one MCP server plus per-agent loadouts. You toggle skills and tools per agent, and the agent only sees what was chosen. This gives you human-in-the-loop control over blast radius without rebuilding configurations per tool.

Question 4

Is SkillzBench available now, and what will it cost?

Accepted Answer

SkillzBench is pre-launch and actively being built; the waitlist/newsletter is live today. Planned pricing includes a generous free tier, Pro (~$9-15/mo) for more agents and unlimited enabled skills, Teams (~$15-25/seat/mo) for org loadouts, allowlists and audit, and Creator (~$5-10/mo) for verified authors, plus benchmark-as-a-service for priority runs.

Question 5

How is SkillzBench different from just pulling skills straight from GitHub?

Accepted Answer

Raw GitHub listings give you no version pinning, no security scan, no license check, and no evidence a skill helps. SkillzBench continuously crawls and dedupes skills from GitHub, then pins each version to a commit SHA, scans it, benchmarks it in a sandbox, and scores it 0-100 — and keeps tracking upstream changes with re-scans, re-benchmarks and notifications after you adopt it.

Verified Agent Skills for Sales Teams