Question 1

What exactly does SkillzBench measure about an Agent Skill?

Accepted Answer

SkillzBench benchmarks every skill in a sandbox using the same model, run both with and without the skill, and publishes the Lift in percentage points. Lift is always displayed separately from the 0-100 trust score, which is built from security, quality, license, maintenance, and community signals — and money never touches the scores.

Question 2

How does SkillzBench handle security risks in SKILL.md packages?

Accepted Answer

Every skill goes through the Verify pillar: it's pinned to a commit SHA, validated, license-checked, security-scanned, and quality-judged. This matters because published research found ~of skills carry at least one vulnerability. At launch, scope is instruction-only skills (SKILL.md plus text references); script-bearing skills are listed link-out only.

Question 3

Which agents and tools does SkillzBench support?

Accepted Answer

SkillzBench targets Agent Skills — SKILL.md instruction packages — used with Claude Code, Codex, Cursor, Gemini CLI, and chat-app agents. Distribution happens through one MCP server plus per-agent loadouts, where you toggle skills and tools per agent and the agent only sees what was chosen.

Question 4

When can I start using SkillzBench?

Accepted Answer

SkillzBench is pre-launch and actively being built — the waitlist and newsletter are live now, but the product itself is not yet public. Joining the waitlist gets you notified as each pillar (Discover, Verify, Benchmark, Equip, Stay current) comes online.

Question 5

How is SkillzBench different from just browsing GitHub for skills?

Accepted Answer

GitHub gives you raw SKILL.md files with no verification, no benchmarking, and no way to tell a good skill from a risky one. SkillzBench dedupes and verifies every skill, security-scans it, pins it to a commit SHA, and publishes measured Lift from paired sandbox testing — plus it tracks upstream changes and re-benchmarks when skills change.

Agent Skill Benchmarking for Designers