Skip to content
ReelQuoteAI logo

ReelQuoteAI MCP server

Which servers in the official MCP registry are ReelQuoteAI's own, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-21

← Back to ReelQuoteAI

Whose server it is
  • Registry namecom.reelquoteai/survey-toolverified in the MCP registry

The registry name sits under reelquoteai.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-21
  • Stateanswerschecked 2026-10-21
  • Tools32
  • Server calls itselfreelquoteai-us
  • Server version1.1.0
  • Protocol version2025-06-18

On 2026-10-21 the server answered initialize and tools/list without a sign-in. First answer in 71 ms.

Should an agent connect itNot proven either way yet
  • Whether its tools do the product's job has not been judged yet.
  • Checked on 3 of the 7 days it takes to judge reliability.

Once connected

  • For "remove items customers say are staying from the inventory", the likely first call is edit_items.
  • Keep approval on for send_resume_link, revoke_key: they delete, send or pay without saying so.
  • create_job, set_contact, create_upload_url, complete_upload, edit_items change data without a sign-in: anyone with the address can call them.
  • Parameters are untyped or undescribed in 10 of 32 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 3, the first on 2026-10-21
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 3 so far
  • Answeredmeasuring3 of the 7 daily checks it takes to print a share
  • Median first answer119 ms
  • Space in an agent's context≈ 5,691 tokensthe definitions of all its tools (32), which an agent carries on every turn
  • Tool definitionsA · 95 of 100
  • Typed results0 of 32 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 10 of 32

What an agent can do with it

  • not sure · turn customer phone walkthrough videos into itemised inventory lists
  • not sure · quote moving jobs from a shared inventory using your own rates
  • can do · remove items customers say are staying from the inventory

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

  • Tools that delete, send or pay without telling the client so: send_resume_link, revoke_key
  • Tools that change data, on a server that answers without a sign-in: create_job, set_contact, create_upload_url, complete_upload, edit_items

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (30%), doing the product's job (25%), definitions (20%), context (10%) and use (15%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers32 tools · read 2026-10-21
  • get_session

    Resolve the key to its mover (name, branding, config). Scope: quote.

  • list_home_profiles

    Home sizes for the ballpark step (2br_apt, 3br_house, …). Scope: quote.

  • ballpark_estimate

    Instant estimate range from home size + addresses. Free; never bookable. Scope: quote.

  • create_job

    Start a job (the survey container). Contact details are NOT accepted here — funnel rule. Send `Idempotency-Key` so a retry cannot create a second job. Scope: jobs:write.

  • list_jobs

    List jobs for the tenant, newest first. Secret key only. Paged: pass `limit` (1–100, default 25), optional `status`, and `cursor` from the previous page's `next_cursor` (null on the last page; a bad cursor is a 400). Scope: jobs:read.

  • get_job

    Full job state: status, inventory, quote, access, terms, move_date. Scope: jobs:read.

  • set_contact

    Store customer email + optional name + move date. Sends NOTHING — required before upload-url. Scope: jobs:write.

  • create_upload_url

    Presigned upload for the walkthrough video. Requires stored contact email. Body: { content_type, consent, size }. Send the byte size: at or above 16 MB the response carries `multipart` (an upload_id, a part_size and a presigned PUT per…

  • complete_upload

    Assemble a chunked upload. Body: { upload_id, parts: [{ part_number, etag }] }. Only after every part has been PUT; then call /submit. Scope: jobs:write.

  • submit_job

    Queue the uploaded video for analysis. Poll GET /jobs/{id} for status. Scope: jobs:write.

  • hand_off_video

    Server-side handoff: give us a public https URL of the walkthrough instead of uploading it from a browser. Secret key only. We fetch it (up to 500 MB, streamed) and the job runs exactly as an upload would — poll GET /jobs/{id}. Needs the…

  • confirm_access

    Confirm floors / elevator / stairs / carry distance for both ends. Scope: jobs:write.

  • edit_items

    Edit inventory (quantity, exclude, add by item_key). Response carries the re-priced quote. Scope: jobs:write.

  • approve_inventory

    Customer sign-off on the inventory, plus free-text notes. Scope: jobs:write.

  • acknowledge_gaps

    Record the customer's answer to the capture-gap prompts (inventory.gaps on GET /jobs/{id}). Body: { acknowledged: true } stamps "nothing is missing"; false clears it. The prompts themselves are recomputed from the current inventory on…

  • send_resume_link

    Email the customer their resume link. Only on explicit request. Scope: jobs:write.

  • request_booking

    Request the booking. Idempotent per job; requires approve:true (types-name-as-signature approval of the inventory + quote, stored with a quote snapshot) and accept_terms when the mover has terms. Scope: jobs:write.

  • report_outcome

    Report actuals (hours, final cents, crew). Secret key only. Scope: outcomes.

  • get_cube_sheet_csv

    Itemized cube sheet as CSV (Excel-ready): rooms, volumes, weights, exclusions. Secret key only. Scope: jobs:read.

  • get_date_prices

    This job's quote priced across a ~2-week window from the mover's date rules. flat=true means every day prices the same. Scope: jobs:read.

  • set_move_date

    Set or clear the move date; invalidates and recomputes the quote. Refused after a booking is requested. Scope: jobs:write.

  • quote_inventory

    Direct pricing for agents: items or total_cuft in, firm quote out. Secret key only. Scope: quote.

  • get_availability

    Which days the mover has marked off, before a job exists. Soft: a marked day can still be requested — the mover confirms and calls first — so steer to an open day when the customer has a choice. Scope: quote.

  • list_keys

    List the account's API keys — live and test, active and revoked. Never a raw secret. Secret key with the `keys` scope. Scope: keys.

  • create_key

    Mint a key. The raw secret is in THIS response only. A key can never mint one broader than itself; a test key mints only test keys; a live secret key may mint the sandbox pair. Scope: keys.

  • rotate_key

    Replace a key with its twin (same kind, mode, scopes, label). The old key keeps working until `replaced.expires_at` so a deploy can swap values without a gap. Needs the `keys` scope even to rotate itself — rotate a scoped key with a full…

  • revoke_key

    Revoke a key immediately. The only live publishable key is refused (409) — rotate it instead. A test key cannot revoke a live key (403 `mode_mismatch`). Scope: keys.

  • list_webhook_deliveries

    The delivery log: every webhook POST for 30 days, newest first, with its outcome and the exact body sent. Secret key with the `keys` scope. A key reads the deliveries of its own mode: live keys the live log, test keys the sandbox log…

  • replay_webhook_delivery

    Re-send one delivery to its endpoint: the same body byte for byte (same envelope id, so your dedupe still works) under a fresh signature, with the usual retries. Returns the replay's own delivery record. Deliveries of the other mode are…

  • search_catalog

    Search the cube-sheet catalog (?q=sofa). Scope: quote.

  • suggest_addresses

    Address autocomplete (proxied; send session_token). Scope: quote.

  • resolve_address

    Resolve a suggestion to a full address (same session_token). Scope: quote.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "survey-tool": {
      "type": "http",
      "url": "https://widget.reelquoteai.com/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http reelquoteai https://widget.reelquoteai.com/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://widget.reelquoteai.com/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.