Decision summary
Overview
What is SkillsTested?
SkillsTested is a beta directory of tested Claude Code skills. A visitor describes an outcome and the site builds a "tested skill path": the order of skills, every handoff, and a label for each stage of the evidence. Current paths include Landing page to visual QA (partial evidence), Screenshots to promo video (partial evidence) and Notes to Obsidian vault (needs retest). The site reports 8 skill packages tested, 22 individual skills installed, 2 failures in package reports and a 2.5-second median install time, with the latest evidence dated 2026-07-29. Each listing leads to a test record and a reproduction pack, and the site publishes head-to-head comparisons of competing skills. The beta only matches published paths and refuses unsupported goals; the goal stays in the browser. The figures are the site's own, and the tested set is small.
Track and trend
- Track: Directory and evaluation of AI agent skills (Claude Code, Codex, Cursor), a curation niche inside the agent-tooling ecosystem.
- Trend it rides: Rides the Claude Code skills and agent-skill-marketplace wave, offering "tested" as the differentiator against untested registries. Its evidence is partial and dated, and the tested set is a handful of packages.
Main features
- Goal to skill path: Describe an outcome and get an ordered path of tested skills with handoffs.
- Evidence labels: Each stage is labelled partial evidence, needs retest or verified.
- Reproduction packs: Inspect the test record and download a reproduction pack before installing.
- Head-to-head comparisons: Two competing skills get the same brief and are compared side by side.
Pricing
Free to use. The site says every skill it has tested is free, open-source Markdown; costs appear at use time because some skills drive paid models or need heavy runtimes, which the test records flag per skill.
Similar tools on AIGCLIST
- Agent Skills: Community registry for discovering, evaluating and installing agent skills for Claude Code.
- LaunchIntel: Curated, categorised directory playbooks for Claude Code or Codex, each tested.
- DSH Hub: Focused directory of DeepSeek Harness plugins and clients.
Limitations
- Only 8 packages and 22 skills tested so far.
- The beta refuses goals outside its published paths.
- Evidence is partial and dated 2026-07-29.
- Test figures are the site's own.
Summary
A small but honest curation site whose main value is its evidence labels. Use it as a starting point, read the test records, and expect it to cover only a few goals for now.
Reviews (0)
No reviews yet. Be the first to rate this product!
Score anatomy
The dimensions behind the editorial score, each with its judgment note. AI Readiness and GEO Score are platform assessments generated by AIGCLIST after submission.
Evidence check
Public claims about this tool, each tagged with a verification status and its cited source.
Decision desk
The questions most worth resolving before you rely on the product or visit its official site.
Continue exploring
More in AI Agents Directory
Published tools that share this product's primary category. They are discovery links, not editorial comparisons.