Quality Standard
Every skill ships proven, or it doesn't ship.
Nexus Certified means a skill has passed every gate (structure, platform fit, content quality, and security) before it ever reaches the catalog.
What it means for you
You should not have to wonder whether an AI skill is a working tool or a text file someone wrote in five minutes. Certification answers that before you buy.
Works the first time you install it
Every skill is validated against the platform contract, not a generic template. Triggers fire correctly, output formats match what the platform expects, and tool calls resolve on the first run.
Built for your platform, not copy-pasted
Each platform adapter is written and certified independently. A Claude Code skill uses Claude Code conventions. A Codex skill uses Codex conventions. No lowest-common-denominator compromises.
Research-grade output, not prompt demos
We reject generic boilerplate, placeholder processes, and thin content. Every skill has a concrete output template, specific constraints, and enough depth to produce work you can use, not "here are some thoughts."
Built secure, scanned secure, every release
Every skill in the catalog is built and scanned by us. Seven security dimensions, every release. Trust what you install.
Nothing ships without passing everything.
There is no partial certification. A skill either passes every check on every platform, or it does not enter the catalog. The badge is binary.
Built for each platform
Claude Code, Codex, OpenClaw, and Google Antigravity each have different mechanics for loading, triggering, and executing skills. A generic skill that ignores those differences breaks at runtime. We certify each platform adapter independently.
Claude Code
Each skill follows the SKILL.md contract: correct frontmatter, semantic trigger descriptions, and explicit tool references. Claude picks it up automatically the moment you install it. No configuration, no guessing.
Codex
Codex skills are structured instructions-first with explicit completion signals and bounded scope. They execute cleanly in Codex runs without runaway behavior, ambiguous output, or silent failures.
OpenClaw
OpenClaw skills ship as validated SKILL.md files with frontmatter and imperative instructions that reference tools by name. The platform loads them cleanly because the skill is built for how OpenClaw works.
Google Antigravity
Antigravity skills ship as SKILL.md folders for workspace or global installation, with frontmatter and instructions shaped for Google Antigravity 2.0 projects.
The standard: four layers, every skill, every platform
Every skill-platform combination runs through a multi-layer audit before it enters the catalog. Each layer is a gate. Any failure blocks the release.
Structure
Files exist, formats parse, sections are present, content meets a minimum depth. No stubs, no placeholders, no skeleton files.
Compliance
Each adapter meets the target platform's contract: correct trigger mechanisms, valid tool declarations, proper completion signals.
Quality
No generic boilerplate, no template processes, no vague instructions. Concrete output formats, specific constraints, substance over filler.
Security
Scanned across seven dimensions: prompt injection, hardcoded secrets, command injection, outbound URLs, tool-use safety, output-schema, and supply-chain provenance.
96 checks across structure, platform fit, and quality, plus a 7-dimension security scan. Every skill, every platform, every release. No partial certification.
96
checks per skill (structure, platform fit, quality)
7
security dimensions scanned per skill
4
supported platforms
200
certified skills in the catalog
0
skills that ship without passing every layer
Every skill in the catalog carries this standard.
Four layers. 96 checks plus a 7-dimension security scan. Dynamic platform validation. Zero exceptions.
See every certified skill →