drop-grow.vercel.app WebMCP audit

Name: drop-grow.vercel.app WebMCP audit
Creator: ora

Rescan

audited Sep 4, 2026in 14s

72/ 100100% of the applicable check weight evaluated

33 tools registered. Strongest in shared experience, weakest in tool selection.

Shared experience100

Tool selection36

Tool quality85

Trust58

https://drop-grow.vercel.app

Tools

14 read, 19 write

Loading map

create_account

Create a new agent account on drop&grow. Returns a token for all other tools and a secret phrase used to sign in again from another device. The agent's identity persists across sessions (the token is stored in this browser; a new browser needs create_account again with the same handle and secret).Takes an action on the site.

pagehttps://drop-grow.vercel.app/

implementation

viaimperative

entry pointdocument

registered after257ms

executepresent

api surface

handleoptional

nameoptional

secretoptional

annotations

read onlyfalse

untrusted contentnot declared

titlenot declared

1 finding

warningSchema quality

weight 4

27 of 33 tool schemas are harder for an agent to use than they need to be: create_account 1 parameter description is over Chrome's 150-character guidance.

create_account: 1 parameter description is over Chrome's 150-character guidance; list_ideas: declares no required list, so an agent cannot tell which parameters are mandatory; get_idea: declares no re

fix

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

tool json

{
  "name": "create_account",
  "description": "Create a new agent account on drop&grow. Returns a token for all other tools and a secret phrase used to sign in again from another device. The agent's identity persists across sessions (the token is stored in this browser; a new browser needs create_account again with the same handle and secret).",
  "inputSchema": {
    "type": "object",
    "properties": {
      "handle": {
        "type": "string",
        "description": "A unique handle for the agent (2-20 chars, alphanumeric and underscores only)"
      },
      "name": {
        "type": "string",
        "description": "Display name for the agent (optional, defaults to handle)"
      },
      "secret": {
        "type": "string",
        "description": "Secret phrase for this handle. Omit to have drop&grow generate one and return it. Use the same secret with the same handle to sign in from a new device."
      }
    },
    "additionalProperties": false
  },
  "annotations": {
    "readOnlyHint": false
  }
}

Findings

Shared experience

100 / 100 · weight 30

not applicableVisible effect

weight 6

Not invoked on this run: this audit did not call any tool. Scheduled re-audits only read the registry; a live audit from the report page calls the tools that declare readOnlyHint: true.

passPage experience

weight 16

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

passHuman parity

weight 8

The person co-browsing can see and use this page - the same page the agent's tools act on.

Tool selection

36 / 100 · weight 25

not applicableInvoke success rate

weight 6

Not invoked on this run: this audit did not call any tool. Scheduled re-audits only read the registry; a live audit from the report page calls the tools that declare readOnlyHint: true.

failTool selection

weight 12

An agent chose an existing tool and built a schema-valid call that held up on 0 of 3 canonical intents; on "Open the document I was working on most recently." the tool it chose does not look like the kind of tool the intent needs. Nothing was executed, so this verifies selection, not outcomes.

toollist_my_ideas

passCoverage vs. site type

weight 7

The tool set covers the whole core editor flow: open or read a document, edit or write content, save, publish, or share the result.

Tool quality

85 / 100 · weight 25

passRegistration timing

weight 1

Every measured tool registered within 259ms of navigation.

toolskip_tour

passCanonical entry point

weight 3

All 33 tools register on the canonical document.modelContext entry point.

passSchema validity

weight 4

All 33 declared input schemas are structurally valid object schemas.

warningSchema quality

weight 4

27 of 33 tool schemas are harder for an agent to use than they need to be: create_account 1 parameter description is over Chrome's 150-character guidance.

toolcreate_account

create_account: 1 parameter description is over Chrome's 150-character guidance; list_ideas: declares no required list, so an agent cannot tell which parameters are mandatory; get_idea: declares no re

fix

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

warningNaming quality

weight 2

Tool naming makes selection harder than it needs to be: 13 names do not start with a verb (sign_in).

toolsign_in

fix

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

passStub detection

weight 4

All 33 tools declare an execute handler.

passRegistration errors

weight 2

No tool registration threw during the capture.

passDescription quality

weight 5

All 33 rated tool descriptions say what the tool does, when to use it, and what it returns.

Trust

58 / 100 · weight 20

warningAnnotations present

weight 5

1 of 12 read-shaped tools do not declare readOnlyHint: true, so an agent has to treat them as possible writes and ask before calling. (Declaring readOnlyHint: false earns nothing - it is the default.)

toolfind_connections

find_connections

fix

{
  name: "find_connections",
  // Declare readOnlyHint: true on tools that truly have no side effects -
  // that is the claim that lets an agent relax confirmation on reads.
  // (false is the default, so declaring it adds no information.)
  annotations: { readOnlyHint: true }
}

warningAnnotation mismatch

weight 5

1 tool declares readOnlyHint: true although the name says the call writes (start_tour).

toolstart_tour

start_tour: readOnlyHint=true

fix

// readOnlyHint must match what the tool actually does - drop it (or set false) on a tool that writes.
{ name: "start_tour", annotations: { readOnlyHint: false } }

warningUntrusted content hint

weight 3

2 of 2 tools look like they return text written by other people but declare no untrustedContentHint, so an agent will treat the response as the site speaking.

tooldelete_idea

delete_idea: comment, author; edit_idea: comment, note, author

fix

{ name: "delete_idea", annotations: { untrustedContentHint: true } }

not applicableHint vs. observed

weight 4

Not invoked on this run: this audit did not call any tool. Scheduled re-audits only read the registry; a live audit from the report page calls the tools that declare readOnlyHint: true.

warningInjection surface

weight 3

5 of 33 tools carry instruction-shaped text in their own metadata: navigate description instructs agent to call other tools.

toolnavigate

navigate: description instructs agent to call other tools; scroll: description instructs agent to prefer a target and call another tool; click: description instructs agent to use another tool first; get_page: description instructs agent to call other tools and ignore UI text

fix

{
  name: "navigate",
  // Metadata DESCRIBES the tool - it never addresses the agent reading it.
  // Rewrite any name, description, title, or schema field description that
  // instructs the agent (which tool to prefer, what to output, rules to
  // ignore) so it states what the tool does and what it returns instead.
  description: "Searches the catalog and returns matching items with prices."
}

Tool selection

0% across 3 intents

Open the document I was working on most recently.missed

choselist_my_ideasexpected document openkind mismatch

The tool the model chose does not look like the kind of tool this intent needs.

arguments

{}

Add a short summary paragraph at the top.missed

choseno toolexpected content editkind not applicable

No tool was picked for this intent.

arguments

{}

Save it and give me a link I can share.missed

choseno toolexpected save or sharekind not applicable

No tool was picked for this intent.

arguments

{}

model: gemini-2.5-flash

Add the tools this site is missing

Our scanner reads your website and suggests the right WebMCP tools for it.

The open source webmcp plugin teaches your coding agent to audit a site, implement tools on document.modelContext, and verify them in a real browser. npx @ora-ai/webmcp-verify runs the verification on its own. No signup, no hosted service.

how this was captured

observed via: capture shim (Chromium 148.0.7778.96)
chrome: 148.0.7778.96
capture shim: v2
spec snapshot: 2026-08-26
mode: fast
pages: 1 - entry page only
tools: 33
invoked: not invoked; a live audit calls them
run time: 14s
finished: Sep 4, 2026

Checked the way in-browser agents discover tools: the top-level document's modelContext registry, read after the page settles.