spectral

I've been diving into API linting tools lately, and I'm curious how you decided on the architectural approach for Spectral—the way it handles rule composition feels like it solves a real pain point in the OpenAPI validation space.

Links:

• GitHub
• Your agentic skill in the SkillzWave marketplace

The TL;DR

You're at 95/100, solidly in A-territory. This is based on Anthropic's best practices for Claude skills. Your strongest area is Writing Style (10/10—the documentation is genuinely clear and dense without being fluff), but you've also nailed Progressive Disclosure Architecture (28/30). The weakest spot is Spec Compliance (12/15), mostly around trigger phrase coverage, but that's a quick fix.

What's Working Well

• Rock-solid structure: The SKILL.md overview with all heavy lifting pushed to references/ is textbook PDA. References for JSONPath, custom functions, and built-in functions sit exactly one level deep—no navigation chaos.
• Token efficiency is genuinely impressive: Every sentence earns its place. No marketing fluff, no "this is great because," just pure information density. That matters a lot for agent prompts.
• Triggers are targeted: "API linting," "custom rulesets," "validating"—these activate appropriately when someone needs to validate OpenAPI specs or set up custom validation rules.

The Big One: Missing Trigger Phrases

You're leaving discoverability on the table here. Your description pulls in 1-2 trigger phrases, but for a tool as versatile as Spectral, you should have more coverage. Right now someone looking for "AsyncAPI validation" or "OpenAPI linting" might not find you.

Fix: Expand the description triggers to include AsyncAPI, OpenAPI, ruleset validation, API specification, and linting rules. This isn't just vanity—it's about meeting develo...