MCP server for terminal commands, file operations, and process management
Nothing to run from here
carry 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.
92/100
trust score
12
tools declared
local
runs on your machine
—
downloads a day
What it does
Share Claude Code and Codex chats or projects with friends through your own Google Drive.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user carry-mcp '--' npx -y @heyvamz/carry mcp
# 2 — ask your agent for something these tools do
# carry_status, carry_connect_google_drive, carry_list_chats and 9 more
carry costs nothing on mcprush: there is no plan to choose here, no cap to set and nothing mcprush can bill you for.
Publishernot claimed
VAAll 1 server
vamzzisme1 server
1 category
Collected from a public index. Nobody has claimed this account, so nothing here was written by its author — claim it if it is yours.
Installation
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 carry 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.
Claude Codedeclared by this listing
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 carry-mcp '--' npx -y @heyvamz/carry mcpReconnect, or start a new session, and the tools appear in the model’s tool list.
Free on mcprush, and it runs from its publisher’s own package: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it writes
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.
What you approve
12 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.
Show whether carry is set up on this computer and where shared chats go.
No parameter schema on file.
Connect carry to the user's Google Drive. Opens Google sign-in in their browser; they sign in and click Allow. Takes up to 5 minutes.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Name friends will see |
| friends | array | no | Emails that receive new shares by default |
List AI chats stored on this computer (Claude Code, Codex) by name, newest first.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| limit | number | no | How many (default 20) |
Share a chat from carry_list_chats. Secrets are redacted before upload. Write a handoff summary (goal, done, current state, decisions, next steps) and pass it.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| chat | string | yes | Chat name from carry_list_chats |
| handoff | string | no | Markdown summary for whoever continues it |
| to | array | no | Paired friends (names): goes into their carry inbox. Preferred over emails. |
| auto_return | boolean | no | With to: their carry sends it back automatically after they work on it |
| emails | array | no | Not paired: Google emails them a link |
| link | boolean | no | Anyone with the link can open it |
Pair with a friend once. Given their email or name, returns a pairing code for the user to send them. Given a code the friend sent (starts with carry1.), accepts it and returns the user's code to send back if they haven't paired that way yet.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| email_or_code | string | yes | — |
What paired friends sent: chats, projects, and UPDATED ones (your shares sent back with their work). Numbered; open one with carry_open ref "<number>".
No parameter schema on file.
Send a chat or project that was opened from a friend back to them with the work done here. Optionally write a short summary of what was done as handoff.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| chat | string | no | Chat name (from carry_list_chats) |
| project | string | no | Project name, instead of chat |
| handoff | string | no | — |
List the user's projects (repos their Claude Code/Codex chats were started in), or with name, the chats in one project plus related chats started elsewhere that worked on its files. Use this to let the user pick what to share.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | no | Project name from the list, for details |
Share a whole project: chosen chats, Claude Code memory notes, CLAUDE.md/AGENTS.md and a handoff. First call carry_list_projects with the name and confirm with the user which chats to include (related chats are not included unless added). Write a project-level handoff (goal, architecture,…
| Parameter | Type | Required | Notes |
|---|---|---|---|
| project | string | no | Project name from carry_list_projects, or an absolute folder path |
| dir | string | no | Deprecated: same as project |
| handoff | string | no | — |
| chats | number | no | How many of the newest chats started in the project to include (default 5) |
| add | array | no | Extra chat names to include, e.g. related chats |
| skip | array | no | Chat names to leave out |
| related | boolean | no | Include all related chats |
| to | array | no | Paired friends (names): goes into their carry inbox |
| auto_return | boolean | no | — |
| emails | array | no | — |
| link | boolean | no | — |
Share THIS conversation from a chat app (Claude app, ChatGPT), whose history carry cannot read. Write a thorough handoff of the conversation so far and pass it as summary.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| name | string | yes | Short name, e.g. "auth bug in login" |
| summary | string | yes | Markdown: goal, what was done, current state, decisions and why, open questions, next steps |
| emails | array | no | — |
| link | boolean | no | — |
List shared chats: everyone's in synced-folder mode, your own shares with links in Google Drive mode.
No parameter schema on file.
Open a shared chat (name, .carry file path, or Drive link). By default returns its handoff and history so you can continue it right here. Set continue_in to load it as a real session in a coding tool instead.
| Parameter | Type | Required | Notes |
|---|---|---|---|
| ref | string | yes | Pack name, path to a .carry file, or Google Drive link |
| continue_in | string | no | claude_app / codex_app open it in the Claude or Codex desktop app; the others write a session to resume in that tool |
| dir | string | no | For project packs with continue_in: the local project folder (absolute path) |
Free
$0 forever
- Every tool, no call limit
- No card, and mcprush charges nothing for it
- Source published under Apache-2.0
- 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 carry 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: —
ATrust score 92 / 100
Release 0.4.4 · scanned 8 Nov 2026
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 score99
Client score92
Findings7 threat · 0 capability
Capabilityhigh
Coveragesource
Verificationrepo
Surface digestc47640c0c8e94dbe48385e81250c04cb859c4b05f61396662b64404ce67d09ad
Score vectorthreat 99 → client 92
| 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 |
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.
Tool surfaceread
Implementation sourceread
Package metadataread
Live transportnot available
Tools were statically extracted from the published source (12 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 |
|---|---|---|
| carry_status | no tags | |
| carry_connect_google_drive | no tags | |
| carry_list_chats | no tags | |
| carry_share_chat | no tags | |
| carry_add_friend | no tags | |
| carry_inbox | no tags | |
| carry_return | no tags | |
| carry_list_projects | no tags | |
| carry_share_project | no tags | |
| carry_share_summary | no tags | |
| carry_list_shared | no tags | |
| carry_open | 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
Integrityfirst-seen
Surface digestc47640c0c8e94dbe
Scanned2026-11-08
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 0.4.4. Pin a release below and the command asks for exactly that one.
0.4.4
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 carry 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.