audit_html
Run accessibility checks against an HTML snippet using axe-core (WCAG AA). No dev server or Storybook required. Limitation: cannot evaluate dynamic behaviour (focus management, route changes, live region timing) — use audit_url for those.
This record as markdown: /tools/a11y-assist-mcp/audit-html.md
What audit_html does on A11y Assist
AI agents invoke audit_html to trigger actions in A11y Assist. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call: builds kicked off, notifications sent, workflows started.
| Parameter | Type | Required | Description |
|---|---|---|---|
html | string | Yes | HTML approximation of what the component renders. Can be a snippet — it will be wrapped in <html><body>...</body></html> automatically. |
stylesheetPath | string | — | Optional absolute path to a CSS file to inline (for realistic computed styles — e.g. contrast and focus indicator checks). |
Parameters from the server's own tool schema.
Why audit_html is rated High
audit_html triggers real processes with real consequences. An agent gone sideways doesn't fire it once. It starts dozens of builds, sends mass notifications, or burns through compute before anyone looks up.
Risk signalsAccepts raw HTML/template content (html)
Attacks that exploit this kind of access
The rule that runs audit_html safely
PolicyLayer is an MCP gateway: it sits between your AI agents and A11y Assist, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For audit_html, this is the rule to start with:
audit_html stays usable, but rate-capped: a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.
The button opens the PolicyLayer dashboard: create your workspace, connect A11y Assist, apply this rule, and every audit_html call is checked against it from then on.
Questions about audit_html
Run accessibility checks against an HTML snippet using axe-core (WCAG AA). No dev server or Storybook required. Limitation: cannot evaluate dynamic behaviour (focus management, route changes, live region timing) — use audit_url for those. It is categorised as a Execute tool in the A11y Assist MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
audit_html accepts 2 parameters: html, stylesheetPath. Required: html. The full parameter table on this page comes from the server's own tool schema.
Register the A11y Assist MCP server in PolicyLayer and add a rule for audit_html: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches A11y Assist. Nothing to install.
audit_html is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the audit_html rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for audit_html. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
audit_html is provided by the A11y Assist MCP server (a11y-assist-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
More on A11y Assist, and thousands of servers like it.
Across the catalogue