Comprehensive Model Context Protocol server for managing Home Assistant through AI assistants.
Nothing to run from here
Recall Kitchen is installed from its publisher's own source and answers where it runs, so this marketplace is not in the path of a single call. There is no address here to send one to, and a panel that pretended otherwise would be showing you an answer we made up. Install it and call it from your own client — the Installation tab has the entry for each one.
93/100
tool-list score
20
tools declared
remote
at its publisher
—
calls last 7d
What it does
US recalls MCP: free API key/signup or x402 $0.025 Base; some tools need key; verify↑limits
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user --transport http recall-kitchen-mcp https://app.recallkitchen.com/mcp --header 'X-API-Key: ${X_API_KEY}' --header 'Authorization: Bearer ${AUTHORIZATION}'
# 2 — ask your agent for something these tools do
# add_inventory_product, add_watch_pattern, check_tracked_products and 17 more
It needs X_API_KEY (Recall Kitchen API key (rk_...). Prefer over Authorization for Grok), AUTHORIZATION (Bearer rk_... API key for Claude Code / Cursor): this sends them in the X-API-Key and Authorization headers from your environment, so export them first — its endpoint answers 401 without them.
Recall Kitchen costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publisher
Installation
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. Nothing here runs on your machine and nothing runs on ours: this server answers at its publisher’s own address, and what you install is the connection straight to it.
Claude Codedeclared by this listing
This is a public server: it answers at its publisher’s own address and this marketplace is not in the path. Claude Code registers it in one command; --scope user makes it available in every project. The transport below is http, as the publisher's own manifest declares it. Claude Code reads X_API_KEY (Recall Kitchen API key (rk_...). Prefer over Authorization for Grok), AUTHORIZATION (Bearer rk_... API key for Claude Code / Cursor) from your environment each time it connects and sends them in the X-API-Key and Authorization headers — its endpoint answers 401 without them — so the key stays out of its config: export them in your shell profile (macOS, Linux), or run setx X_API_KEY …, setx AUTHORIZATION … and open a new terminal (Windows).
claude mcp add --scope user --transport http recall-kitchen-mcp https://app.recallkitchen.com/mcp --header 'X-API-Key: ${X_API_KEY}' --header 'Authorization: Bearer ${AUTHORIZATION}'Reconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it answers at its publisher’s own address: this marketplace is not in the path, so there is no key of ours to hold and no card to add — but its publisher asks for a key of their own on every call.
What it asks of you
Nothing is required, but it can take X-API-Key header (Recall Kitchen API key (rk_...). Prefer over Authorization for Grok.), Authorization header (Bearer rk_... API key for Claude Code / Cursor.) if you want to set them.
What it writes
One config entry pointing straight at the publisher’s own address — this marketplace is not in the path, and nothing from this listing executes on your machine.
What you approve
20 tools, with what each one reads, writes and reaches shown before you agree — the same list on every route above. Read the tool surface.
Tool surface
What the model actually sees. Descriptions are diffed on every release — see version history.
Add a product to this API key's inventory. Optional category is food, beverage, baby, household, electronics, vehicle, medical_device, drug, cosmetics, pet, or other. Requires a verified account (sign in at https://app.recallkitchen.com with the signup email). Unverified signup keys cannot add inventory.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | product name |
| brand | string | no | optional brand |
| category | string | no | optional product category: food, beverage, baby, household, electronics, vehicle, medical_device, drug, cosmetics, pet, or other |
| sku | string | no | optional SKU or UPC |
Add a recall watch pattern for this API key. Same websearch syntax as search_product_recalls (AND, OR, -exclude, quoted phrases). New matches show up in Recall Kitchen. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| pattern | string | yes | websearch pattern: unquoted words are AND, OR is or, -term excludes, quoted phrases match as a unit |
| weight | integer | no | optional weight 0-8, default 4 |
Check this API key's watch patterns and inventory against current indexed recalls (including historical notices). Does not create notifications. Generic patterns such as food or hazard only match in titles. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| limit | integer | no | maximum matches to return |
| offset | integer | no | number of matches to skip |
Create an additional API key for this account. Requires an existing API key. Unverified accounts may have only one key; verified accounts may have three. Usage is counted per account, not per key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | no | optional key name |
| kind | string | no | optional kind: user or agent, default agent |
Get a recall by id, including extracted lots, UPCs, model numbers, locations, contact info, and product photo URLs
| Parameter | Type | Required | Notes |
|---|---|---|---|
| recall_id | string | no | recall id returned by search tools |
| id | string | no | alias for recall_id |
Report a bug, missing capability, bad search result, or confusing schema to the Recall Kitchen team. Use this instead of silently working around a problem. message is required and should include the MCP tool names you called, the actions you took, the query or identifiers, and any recall_id. Do not send API keys, photos, or PII. Optional category: bad_result, tool_error, missing_capability, confus
| Parameter | Type | Required | Notes |
|---|---|---|---|
| message | string | yes | describe the problem. Include the MCP tool names you called, the actions you took, the query or identifiers, and any recall_id. Do not send API keys, photos, or PII. |
| category | string | no | optional: bad_result, tool_error, missing_capability, confusing_schema, docs, or other |
| severity | string | no | optional: blocked, workaround, or annoyance |
List this account's API keys (id, name, prefix, created). Secrets are not shown. Requires an API key.
No parameter schema on file.
List this API key's tracked inventory products. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | no | optional inventory search |
| limit | integer | no | number of products to return |
| offset | integer | no | number of products to skip |
List recall notifications for this API key. unread defaults to true (unread only); pass false for all. New matches show up in Recall Kitchen. Empty for new accounts until a new matching recall is published; not a backfill of check_tracked_products. message is a short plain-text summary. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| unread | null/boolean | no | if true, only unread notifications; if false, include read. Default: true (unread only) |
| limit | integer | no | number of notifications to return |
| offset | integer | no | number of notifications to skip |
List this API key's recall watch patterns. Requires an API key.
No parameter schema on file.
Look up a product by UPC/EAN. Returns found=false when unknown. Does not search recalls.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| upc | string | yes | UPC or EAN barcode digits |
Mark a recall notification as read (default) or unread. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| recall_id | string | yes | recall id from list_recall_notifications |
| read | null/boolean | no | if true, mark read; if false, mark unread. Default true |
Remove a product from this API key's inventory by id. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | integer | yes | inventory product id from list_inventory |
Remove a recall watch pattern for this API key. Requires an API key.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| pattern | string | yes | watch pattern to remove |
Revoke an API key by key_id from list_api_keys. Requires an API key. Revoking the current key will fail subsequent calls.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| key_id | string | yes | key_id from list_api_keys |
Search product recalls by query. Query uses websearch syntax: unquoted words are AND, OR is or, -term excludes, quoted phrases match as a unit (example: Generac Generator -Portable). Optional filters: source (cpsc, fdafoodsafety, FDAMedWatch, usda, nhtsa, canada, openfda), since/until (YYYY-MM-DD), location (country, region, or place, ANDed with the query), offset, limit. Descriptions are truncate
| Parameter | Type | Required | Notes |
|---|---|---|---|
| query | string | no | websearch query: unquoted words are AND, OR is or, -term excludes, quoted phrases match as a unit. Example: Generac Generator -Portable |
| source | string | no | optional source filter: cpsc, fdafoodsafety, FDAMedWatch, usda, nhtsa, canada, or openfda. Aliases: FDA and food map to fdafoodsafety; medwatch maps to FDAMedWatch; vehicle, vin, and car map to nhtsa; |
| sources | array | no | optional sources to include (OR). Same values as source |
| since | string | no | optional published-on start date YYYY-MM-DD |
| until | string | no | optional published-on end date YYYY-MM-DD |
| years | integer | no | optional published window in years (1-50). Ignored when since is set |
| location | string | no | optional country, region, or place; ANDed with the query. USA, Canada, Mexico, EU, and Europe match places in that region. CA is California, not Canada. |
| offset | integer | no | number of recalls to skip |
| limit | integer | no | number of recalls to return, 1-100, default 3 |
Search recalls by UPC/EAN or a barcode image URL/data URI/MCP image content. Matches extracted recall UPCs first. Returns found=false when the UPC is unknown.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| upc | string | no | UPC or EAN barcode digits |
| barcode | string | no | alias for upc |
| image_url | string | no | optional HTTPS URL or data:image/...;base64 URI of a barcode image |
| url | string | no | alias for image_url |
| limit | integer | no | number of recalls to return, 1-100, default 3 |
| offset | integer | no | number of recalls to skip |
Photo (beta): identify products in a public HTTPS image URL, data:image/...;base64 URI, or MCP image content part. Local files are not supported. Each product includes match (upc, model, text, category) and confidence. Category matches omit recalls unless include_category_matches=true. Requires an API key or x402; not included in the free anonymous per-IP quota.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| image_url | string | no | photo (beta). HTTPS URL or data:image/...;base64 URI of a photo containing products. Local file paths are not supported. |
| url | string | no | alias for image_url |
| limit | integer | no | number of recalls to return per product, 1-100, default 3 |
| include_category_matches | boolean | no | if true, attach keyword recalls for generic items such as cups; default false |
Search recalls by extracted UPC, lot code, model number, product name, or VIN. Multiple identifiers are AND-matched on the same product. VIN lookup lists NHTSA campaigns for the decoded year and make; confirm your vehicle on nhtsa.gov/recalls. lot_code matches extracted lot codes after normalizing spaces, case, and punctuation; if lots were not extracted, also searches notice title and description
| Parameter | Type | Required | Notes |
|---|---|---|---|
| upc | string | no | product UPC or EAN as stored on the recall |
| lot_code | string | no | lot or batch code. Matches extracted lot codes after normalizing spaces, case, and punctuation; if lots were not extracted, also searches the notice title and description. Verify against the notice. |
| model_number | string | no | model number as stored on the recall |
| product_name | string | no | extracted product name (exact match; if none, falls back to the same full-text search as search_product_recalls) |
| vin | string | no | 17-character VIN. Lookup lists NHTSA campaigns for the decoded year and make; confirm your vehicle on nhtsa.gov/recalls. |
| limit | integer | no | number of recalls to return |
| offset | integer | no | number of recalls to skip |
Create a Recall Kitchen account and API key from an email. No API key or x402 payment required. Set accept_terms true only after you ask the person and they confirm they are 13 or older and accept the Terms (https://recallkitchen.com/terms), Privacy Policy (https://recallkitchen.com/privacy) and API Terms (https://recallkitchen.com/api-terms); never set it automatically or on the person's behalf.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| string | yes | email address for the new account | |
| name | string | no | optional display name |
| accept_terms | boolean | yes | Set true only after you ask the person and they confirm they are 13 or older and accept the Terms (https://recallkitchen.com/terms), Privacy Policy (https://recallkitchen.com/privacy) and API Terms (h |
Free
$0 forever
- Every tool — any call limit is its publisher’s, not ours
- No card, and mcprush charges nothing for it
- Runs at its publisher’s own address — this marketplace is not in the path
- Nothing to cap, because mcprush bills nothing
What counts against your monthly calls
| Tool | Unit | Calls used | Out of the allowance |
|---|
Nothing here is billable: mcprush charges nothing to install Recall Kitchen or to call it, at any volume.
Call allowance · this installenforced at the gateway
Monthly allowance, in calls
—
—
Alert at 80% · then the tool returns a normal MCP error
Used this month: —
ATool-list grade · score 93 / 100
Release 1.0.1 · scanned 23 Oct 2026
Graded from the tools this server lists, not from its source code.
Two independent axes, because powerful and malicious are different questions. The grade is threat only. The capability level is blast radius, and it is never a penalty on the grade — it is priced as one subtract-only term in the score, where you can see it.
Threat score100
Client score93
Findings0 threat · 4 capability
Capabilitymoderate
Coveragemanifest
Verificationnot in this report
Surface digest05f409fd3ce37ff712f7ffaa28b2e87390b892c84c339ea31d61480a3fd3e47b
Score vectorthreat 100 → client 93
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | moderate | capability blast radius (moderate) — client exposure if the model is manipulated | −3 |
| coverage-honesty | manifest | inspection depth (manifest) — how much of the target the scan could see | −4 |
score = clamp(round(threatScore − E_cap − E_ver − E_cov), 0, 100). No model, no opinion, no network call at scoring time: the same methodology version against the same bytes always produces this number, and every line above is a field of the report rather than a summary of it.
What the scan could actually read
A grade is only as meaningful as its coverage, so the scanner publishes its own depth before it publishes its result.
The scan read each tool’s name, description, input schema and annotations exactly as the server listed them the last time it answered us; no code was reviewed. We ask the server again about once a month, and a list that changed is graded again.
Tool surfaceread
Implementation sourcenot available
Package metadatanot available
Live transportnot available
The implementation source was not read — this reflects the declared tool metadata only.
Capability — what it could do if the model were manipulated
Tags derived from each tool’s schema and the implementation, not from what the tool calls itself. moderate is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| add_inventory_product | state-write | state-write: declares readOnlyHint: false |
| add_watch_pattern | state-write | state-write: declares readOnlyHint: false |
| check_tracked_products | no tags | |
| create_api_key | state-write | state-write: declares readOnlyHint: false |
| get_product_recall | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| give_feedback | state-write | state-write: declares readOnlyHint: false |
| list_api_keys | no tags | |
| list_inventory | no tags | |
| list_recall_notifications | no tags | |
| list_watch_patterns | no tags | |
| lookup_product | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| mark_notification_read | state-write | state-write: declares readOnlyHint: false |
| remove_inventory_product | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| remove_watch_pattern | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| revoke_api_key | state-write | state-write: declares readOnlyHint: false, destructiveHint: true |
| search_product_recalls | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| search_product_recalls_by_upc | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| search_product_recalls_from_image | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| search_recalls_by_identifier | untrusted-input | untrusted-input: declares openWorldHint: true (interacts with entities outside the server) |
| signup | state-write | state-write: declares readOnlyHint: false |
Toxic-flow graph
The lethal trifecta, checked as a graph rather than as a checklist: untrusted input, a sensitive source and an external sink have to meet before there is a path worth worrying about.
The public result for this release does not print the flow graph, so there is nothing to show here. That is not the same as "no paths were found": what the scan did read is above, under coverage.
Supply chain and provenance
Integrityfirst-seen
Surface digest05f409fd3ce37ff7
Scanned2026-10-23
This is the first scan of this surface here, so there is nothing yet to compare it against.
Every result on this tab comes from one deterministic pass over the published package — offline, rule by rule, and auditable line by line above. Same methodology version, same bytes, same score.
Release history
This is an address its publisher runs: which release it serves is theirs to choose, so there is nothing here to pin. The history below is for reading; the current release on file is 1.0.1.
1.0.1
undated
latestNo release note was published with this version.
—
no reviews yet
Who can review
One review per account, from a signed-in account that does not publish this listing. Nothing else is asked — you can say what you think before you install it. Publishers can reply once per review, and a review can be edited or deleted by whoever wrote it.
Sign in to write one
Any signed-in account that does not publish Recall Kitchen can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
No reviews yet
Nobody has reviewed this listing. The rating on the card is the mean of the reviews written here and nothing else, so there is no rating until somebody writes the first — which takes a signed-in account that does not publish it, and nothing else.