Google Search Console integration for Model Context Protocol (MCP)
Aircheck 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.
What it does
Queues posts as JSON manifests on your machine and re-runs one post's pipeline locally. Never posts.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user aircheck-mcp '--' npx -y aircheck-mcp
# 2 — ask your agent for something these tools do
# queue_post, mark_sent, push_day and 6 more
Aircheck costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Where are you running it?
Every route below installs the same thing and ends at the same approval screen. This one runs on your machine: your client starts Aircheck as a process under your own user, with your files and your network, so the tool surface below is what it can do to your computer rather than to a server somewhere else. Its scan result is on the Security tab — read the surface before you approve it.
This is a public server: you run it yourself and this marketplace is not in the path. You need Node.js — the current LTS (24) is the safe choice; npx comes with it. Claude Code registers it in one command; --scope user makes it available in every project. The '--' is in quotes so that PowerShell passes it on.
claude mcp add --scope user aircheck-mcp '--' npx -y aircheck-mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Nothing is required, but it can take AIRCHECK_CONFIG (Path to aircheck.config.json. Optional.), AIRCHECK_TOKEN (Machine token for an aircheck.dev account. Optional.) if you want to set them.
One config entry your client uses to start the process locally. A local server runs with your file system and your network, which is why it is priced without metering.
9 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.
Read from its source code. A server that builds its tool list when it starts can list more tools than this.
Write one Aircheck manifest for a post you just queued, or an ad creative you just rendered, into <queue>/<YYYY-MM-DD>/<id>.json. The day comes from at in the configured timezone (today when absent). Pass ad to make it an ad creative: same manifest, reviewed in Ads instead of Social. Pass…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| lane | string | yes | Which account/lane this goes out on. Must match the lane id configured in Aircheck. |
| at | string | no | When it sends, ISO 8601. Leave out for an unscheduled post. |
| kind | string | no | text, image, carousel, video, thread or article. Inferred from media when absent. |
| title | string | no | — |
| text | string | no | The copy, exactly as it will post. Not a summary. |
| media | array | no | — |
| url | string | no | The link the post points at. |
| meta | object | no | Anything producer-specific worth keeping with the post. |
| status | string | no | A post: queued, sent, held or failed (default queued). An ad creative: draft, review, approved, live or archived (default draft). A word from the other ladder is refused. |
| ad | object | no | Present only for an ad creative. Its presence is what makes the manifest a creative. lane is then the ad platform or account. |
| object | no | Present only for an email (a newsletter, a product email, a sequence step). Its presence is what makes the manifest an email. lane is then the sender or list. at is when it sends. | |
| id | string | no | The manifest filename. Generated from the lane and the time when absent. |
| day | string | no | Override the YYYY-MM-DD folder. Normally leave this to at. |
After a send, rewrite one manifest: status (default "sent") and postUrl. Every other field is preserved exactly — the file is read, patched and re-serialized.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | no | The manifest id (its filename without .json). |
| path | string | no | The manifest file, instead of an id. |
| day | string | no | Narrow an id to one YYYY-MM-DD folder. |
| postUrl | no | The permalink the platform returned. | |
| status | string | no | Defaults to sent for a post and live for an ad creative. Must fit the manifest: a post takes queued, sent, held or failed; a creative takes draft, review, approved, live or archived. |
Send a whole day to the Aircheck account: every post and ad creative the config's sources hold for that day, read from disk the way the app reads it. Use it when jobs write their own plan files instead of calling queue_post. Safe to repeat; a later push overwrites the earlier copy. Needs…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| day | string | no | YYYY-MM-DD. Defaults to today. |
Read a day's queue back: every manifest, with its media resolved to absolute paths or URLs, and any local file the manifest names that is not on disk. Defaults to today in the configured timezone.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| day | string | no | YYYY-MM-DD. Defaults to today. |
Read the setup and PROPOSE the wiring: where the config is (or exactly what to write when there is none), the posting jobs visible on this machine (cron, launchd, Airflow), a proposed aircheck.config.json, and the change each job needs. It WRITES NOTHING unless apply: true, and even then only the…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| apply | boolean | no | Write the config file. Only after the user has approved that exact config. |
| configFile | string | no | Where to write the config when none exists. |
| queue | string | no | Propose this queue folder. |
| mediaRoots | array | no | Every folder the jobs write media into. Real paths, not symlinks. |
| timezone | string | no | — |
| roots | array | no | Extra code roots to look for Airflow dags under. |
Re-run one queued post's own pipeline locally and return the result. Spawns the configured headless agent in a fresh run folder under dataDir/runs/, told in so many words never to post, publish, schedule, upload or arm anything, and to write only inside that folder. The queued manifest is not…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | no | The manifest id to repair. |
| path | string | no | The manifest file, instead of an id. |
| day | string | no | Narrow an id to one YYYY-MM-DD folder. |
| request | string | no | What is wrong, in the person's own words. |
| agent | string | no | Which agents.commands entry to run. Defaults to agents.default. |
| timeoutMinutes | number | no | — |
| wait | boolean | no | Block until the agent finishes, instead of returning a run id to poll. Only for a repair you expect to be quick. |
| waitSeconds | number | no | How long to block when wait is set. Default 120. |
Call this RIGHT BEFORE you send a queued post. Returns the decision a person made on aircheck.dev (approved, held, changes, or pending) and send: true means post it, false means do not. When the decision is "changes", note says what to fix: fix it, queue the new version with queue_post, and do…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| id | string | no | The manifest id queue_post returned. |
| path | string | no | The manifest file, instead of an id. |
| day | string | no | The YYYY-MM-DD it is queued on. Needed when the manifest is not on this machine. |
Install a macOS login item that runs the Aircheck media bridge on its own, with this server's token and config, so images and fixes on aircheck.dev keep working after this agent session closes. Pass stop: true to remove it.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| stop | boolean | no | Remove the login item instead. |
How a repair run is doing, and its result once it is done. A run whose process is gone is finished from what it left behind, so a result survives a restart.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| run | string | yes | The run id repair returned. |
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under MIT
- Runs on your machine — nothing of it reaches our gateway
- 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 Aircheck or to call it, at any volume.
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.
| Term | Level | What it prices | Points |
|---|---|---|---|
| capability-exposure | high | capability blast radius (high) — client exposure if the model is manipulated | −6 |
| verification-discount | repo | publisher verification (public source) — no provenance, but the source is public and inspectable | −1 |
| coverage-honesty | source | inspection depth (source) — how much of the target the scan could see | −0 |
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.
Tools were statically extracted from the published source (9 recovered), not enumerated from a running server. Tool-poisoning, Unicode-smuggling, capability and toxic-flow analysis ran on this inferred surface, but a mis-parsed registration could be missed or mis-attributed, so tool-derived findings are capped below “confirmed”. To grade the real runtime surface, scan the running server: --command "npx -y <package>".
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. high is the level these add up to.
| Tool | Capability tags | Why the tag was assigned |
|---|---|---|
| queue_post | no tags | |
| mark_sent | no tags | |
| push_day | no tags | |
| list_queue | no tags | |
| configure | no tags | |
| repair | no tags | |
| check_post | no tags | |
| keep_running | no tags | |
| repair_status | no tags |
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
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
Always latest — the install command names no version, so it installs the newest release when it runs; the current release on file is 0.6.0. Pin a release below and the command asks for exactly that one.
No release note was published with this version.
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.
Any signed-in account that does not publish Aircheck can review it, once — before installing it or after. A review from an account that has installed it is marked as one.
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.