audit page

Audit HTML/CSS against Raven's design quality standards. Checks typography (min 13px, weight 400+, modular-scale heading ratios, line-height consistency), acces

liveverifiedunknown~5s typical

mcpprobedraven

$0.0020

per call

plus $0.001 platform fee per run

Open MCP server; Cracked bills routing only

Provider Upstream docs

$Set up https://cracked-ai-pearl.vercel.app/SKILL.md, then use Cracked to run mcp-ai-ravenmcp-raven-mcp /audit_page for me.

Input

Field	Type	Description
html	string	The full HTML content of the page to audit
url	string	REJECTED on this hosted (remote) endpoint; pass html instead. (Local/stdio: Raven launches
scroll_settle	boolean	Before capturing, step through the page with short pauses so IntersectionObserver/whileInV
interactions	array	Before capturing, fire each interaction in order (hover/click/focus the selector, then wai
viewport	object
strict	boolean	Strict mode — also flags warnings as failures. Default: false
containerMaxWidth	number	Your design system's canonical content-container width in px (e.g. 1152). When set, the re
adversarial_verify	boolean	After generating findings, independently re-check each against the live DOM/network and ta
compact	boolean	Return only the decision-grade signal — score, grade, summary, errors, warnings, fix_prior

Call it

curl

curl https://cracked-ai-pearl.vercel.app/v1/run \
  -H "Authorization: Bearer ck_live_..." -H "content-type: application/json" \
  -d '{"provider":"mcp-ai-ravenmcp-raven-mcp","endpoint":"/audit_page","input":{}}'

cli

npx cracked-ai run -p mcp-ai-ravenmcp-raven-mcp -e /audit_page -i '{}'

mcp

run_tool({ provider: "mcp-ai-ravenmcp-raven-mcp", endpoint: "/audit_page", input: {} })

Try it

Runs against your signed-in workspace balance. Sign in if you have not.