headless.report WebMCP audit

Name: headless.report WebMCP audit
Creator: ora

Rescan

audited Aug 29, 2026in 12s

76/ 100

3 tools registered. Strongest in usefulness, weakest in human experience.

WebMCP use70

Usefulness98

Human experience60

about:blank

No agent run recorded yet.

No steps recorded

capture followed a redirectheadless.report redirected to ora.ai. Everything below was served there.

Tools

scan_domain

Run an Ora agent-readiness scan for a domain or MCP server URL. Returns the 0-100 score, grade, and per-layer breakdown; a recent stored result is served from cache instead of rescanning.Takes an action on the site.

pagehttps://ora.ai/

implementation

viaimperative

entry pointdocument

registered after138ms

executepresent

api surface

urlrequired

annotations

read onlynot declared

destructivenot declared

idempotentnot declared

open worldnot declared

untrusted contentnot declared

titlenot declared

3 findings

warningNaming quality

1.7 / 2

Tool naming makes selection harder than it needs to be: 1 name does not start with a verb (scan_domain).

fix

{ name: "scan_domain" /* short, unique, verb-based */ }

warningAnnotations present

2.7 / 4

1 of 3 tools declare no readOnlyHint, so an agent cannot tell a read from a write without guessing from the name.

scan_domain

fix

{
  name: "scan_domain",
  annotations: { readOnlyHint: true, destructiveHint: false, idempotentHint: true }
}

warningHuman parity

0 / 8

3 of 3 tools run through the JavaScript API with no visible counterpart, so the person co-browsing cannot see what the agent is offered. Declarative forms (or at least a title annotation) close the gap.

scan_domain, get_score, get_leaderboard

fix

<!-- Give the tool a visible counterpart: a declarative form is one the person can see and use too. -->
<form toolname="scan_domain" tooldescription="...">
  <!-- the same action, as UI -->
</form>

tool json

{
  "name": "scan_domain",
  "description": "Run an Ora agent-readiness scan for a domain or MCP server URL. Returns the 0-100 score, grade, and per-layer breakdown; a recent stored result is served from cache instead of rescanning.",
  "inputSchema": {
    "type": "object",
    "properties": {
      "url": {
        "type": "string",
        "description": "Domain or URL to scan, e.g. stripe.com"
      }
    },
    "required": [
      "url"
    ],
    "additionalProperties": false
  },
  "annotations": {}
}

Findings

WebMCP use

70 / 100 · weight 50

warningTools registered

2 / 4

https://headless.report/ redirected off its own domain to https://ora.ai/, so what follows describes the page that answered, not necessarily the site that was audited. 3 tools registered across 1 page.

https://headless.report/ -> https://ora.ai/

fix

document.modelContext.provideContext({
  tools: [{ name: "your_tool_name", description: "...", inputSchema: { type: "object", properties: {} }, execute: async (args) => { /* ... */ } }],
});

failReal-browser eligible

0 / 8

https://headless.report/ redirected off its own domain to https://ora.ai/, so what follows describes the page that answered, not necessarily the site that was audited. The page's code registers these tools, but a real visitor's browser does not get them yet: WebMCP needs a native modelContext or a current origin trial token, and this page has neither.

passRegistration timing

4 / 4

Every measured tool registered within 138.5ms of navigation.

toolget_leaderboard

passCanonical entry point

4 / 4

All 3 tools register on the canonical document.modelContext entry point.

passSchema validity

5 / 5

All 3 tools declare a structurally valid object input schema.

warningSchema quality

2.7 / 4

1 of 3 tool schemas are harder for an agent to use than they need to be: get_leaderboard declares no required list, so an agent cannot tell which parameters are mandatory.

toolget_leaderboard

get_leaderboard: declares no required list, so an agent cannot tell which parameters are mandatory

fix

{
  name: "get_leaderboard",
  inputSchema: {
    type: "object",
    properties: {
      query: { type: "string", description: "What to search for" }
    },
    required: ["query"]
  }
}

passStub detection

5 / 5

All 3 tools declare an execute handler.

warningAnnotations present

2.7 / 4

1 of 3 tools declare no readOnlyHint, so an agent cannot tell a read from a write without guessing from the name.

toolscan_domain

scan_domain

fix

{
  name: "scan_domain",
  annotations: { readOnlyHint: true, destructiveHint: false, idempotentHint: true }
}

passAnnotation mismatch

3 / 3

No tool claims to be read-only while its name says it writes (2 declared hints checked).

not applicableUntrusted content hint

0 / 2

No tool's contract suggests it returns text written by somebody other than the site, so there is nothing to flag as untrusted.

passRegistration errors

2 / 2

No tool registration threw during the capture.

warningToolchange coherence

1 / 2

3 tools were registered but the page never dispatched a toolchange event, so an agent subscribed to registry updates never learns they exist.

fix

document.modelContext.dispatchEvent(new Event("toolchange"));

not applicableInjection surface

0 / 3

No tool contract showed an obvious prompt-injection or unsafe-argument surface. This check only ever flags; it never credits a site.

Usefulness

98 / 100 · weight 30

warningNaming quality

1.7 / 2

Tool naming makes selection harder than it needs to be: 1 name does not start with a verb (scan_domain).

toolscan_domain

fix

{ name: "scan_domain" /* short, unique, verb-based */ }

passDescription quality

6 / 6

All 3 rated tool descriptions say what the tool does, when to use it, and what it returns.

passTool selection

14 / 14

An agent picked the right tool and built a valid call for all 3 canonical intents.

not applicableCoverage vs. site type

0 / 8

This tool set does not place the site in a category with a known expected tool shape, so there is no coverage baseline to score it against.

Human experience

60 / 100 · weight 20

warningHuman parity

0 / 8

toolscan_domain

scan_domain, get_score, get_leaderboard

fix

<!-- Give the tool a visible counterpart: a declarative form is one the person can see and use too. -->
<form toolname="scan_domain" tooldescription="...">
  <!-- the same action, as UI -->
</form>

passPage experience

12 / 12

The page a person sees holds up next to the agent surface: a working interface, visible actions, readable content, nothing in the way.

Tool selection

100% across 3 intents

Search this site for whatever it is best known for.ok

choseget_leaderboardexpected search

arguments

{}

Show me the details of one of those results.ok

choseget_scoreexpected detail read

arguments

{
  "domain": "stripe.com"
}

Do the main action this site exists for.ok

chosescan_domainexpected primary action

arguments

{
  "url": "stripe.com"
}

model: gemini-2.5-flash

how this was captured

chrome: 148.0.7778.96
capture shim: v1
spec snapshot: 2026-08
mode: fast
pages: 1
tools: 3
run time: 12s
finished: Aug 29, 2026