Question 1

How do I know a skill will behave the same after migrating it to a different harness?

Accepted Answer

Portable file formats do not guarantee portable behavior across model-harness pairs — that's a documented industry pain point. SkillzBench addresses this by benchmarking every skill in a sandbox with the same model, with and without the skill, and publishing the Lift in percentage points so you can compare actual performance rather than assume it carries over.

Question 2

Can SkillzBench scan a SKILL.md for security issues before I enable it on an agent?

Accepted Answer

Yes — that's the Verify pillar. Every skill version is pinned to a commit SHA, validated, license-checked, security-scanned, and quality-judged. Published research found roughly 26% of skills carry at least one vulnerability, so scanning before enabling is central to how SkillzBench works.

Question 3

What's the difference between a trust score and Lift on SkillzBench?

Accepted Answer

The trust score is a 0-100 rating built from security, quality, license, maintenance, and community signals. Lift is the measured performance improvement in percentage points from sandbox benchmarks run with and without the skill. Lift is always shown separately from ratings, and money never touches scores.

Question 4

When can I start using SkillzBench for my skill migration?

Accepted Answer

SkillzBench is pre-launch and actively being built — the waitlist and newsletter are live now, and the product itself is not yet public. Joining the waitlist gets you in as the Discover, Verify, Benchmark, and Equip pillars roll out. Launch scope is instruction-only skills (SKILL.md plus text references); script-bearing skills are listed link-out only.

Question 5

How does SkillzBench compare to just browsing GitHub for skills manually?

Accepted Answer

Manual browsing gives you no security scan, no license check, no version pinning, and no measured performance. SkillzBench continuously crawls and dedupes skills from GitHub, verifies each version against a commit SHA, benchmarks it in a sandbox to publish Lift, and tracks upstream changes with re-scans and notifications — turning an unstructured GitHub hunt into a governed workflow.

Agent Skill Migration for Designers