MCP server for terminal commands, file operations, and process management
Nothing to run from here
Chat Export 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.
30
tools declared
—
trust score
local
runs on your machine
—
calls last 7d
What it does
Local MCP for eight logged-in LLM UIs via the Chat Export Chrome extension on 127.0.0.1.
Quickstart
# 1 — add it to Claude Code
claude mcp add --scope user chat-export-mcp '--' docker run -i --rm --platform linux/amd64 ghcr.io/svergio/chat-export-server:1.0.4
# 2 — ask your agent for something these tools do
# browsers_list, browsers_bind, browsers_status and 27 more
Chat Export 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. This one runs on your machine: your client starts Chat Export 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. It has not been scanned here, so read the surface, and its source, 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 Docker installed and running. Only the required variables are passed into the container. To set an optional one (CHAT_EXPORT_HOME, CHAT_EXPORT_MCP_TOKEN, CHAT_EXPORT_HOST, CHAT_EXPORT_PORT), add -e NAME before the image name and give it a value in the entry’s env: an unset -e would clear the default the image sets. 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 chat-export-mcp '--' docker run -i --rm --platform linux/amd64 ghcr.io/svergio/chat-export-server:1.0.4Reconnect, 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 image: this marketplace is not in the path, so there is no key of ours to hold and no card to add.
What it asks of you
Nothing is required, but it can take CHAT_EXPORT_HOME (Data directory for SQLite cache, keys and exports), CHAT_EXPORT_MCP_TOKEN (Optional pre-set MCP Bearer token (cem_…); otherwise generated on first start), CHAT_EXPORT_HOST (HTTP bind host (default 127.0.0.1; use 0.0.0.0 in containers)), CHAT_EXPORT_PORT (HTTP port for /mcp, /ws and /health (default 8765)) if you want to set them.
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
30 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.
List Chrome profiles currently connected through the Chat Export extension bridge. Use before browsers_bind to pick a browserId; use browsers_status for the already-bound session only.
No parameter schema on file.
Set the active browserId for subsequent chat/project/job tools. Call after browsers_list; does not open tabs by itself. Prefer browsers_status to inspect the current binding without changing it.
No parameter schema on file.
Show the currently bound browserId, connection state, and recent alerts. Prefer over browsers_list when you already bound a browser and only need health/permissions.
No parameter schema on file.
List supported providers and whether each is enabled on the server and the bound browser. Requires browsers_bind for browser-side enablement; use before chats_* when unsure which platforms are available.
No parameter schema on file.
Probe gate/limit status per provider on the bound browser without sending a prompt. Returns state ok|rate_limit|busy|auth_wall|captcha|generating|composer_blocked|no_tab|offline|error plus code/kind/banner/retryable. Prefer before chat_ask when choosing a platform; set ensureTab:true to open a…
No parameter schema on file.
Open or reuse a provider start tab in the bound browser without sending a prompt. Prefer before chats_list/chat_ask when no platform tab is open; returns tab meta and created=true if chrome.tabs.create ran. Requires browsers_bind.
No parameter schema on file.
Read composer capabilities (models, modes, toggles, tools, attach limits) from the open provider tab. Prefer before chat_ask/chats_send/chats_create when setting model|mode|toggles|tools so ids match the live UI. Use platforms_status for gate/busy/captcha; platforms_list for enablement only.…
No parameter schema on file.
List chats for one platform in the bound browser (titles/ids, optional query). Prefer for discovery; use projects_chats for chats inside one project, chats_get for a full payload, chat_outline for turn indexes. Requires prior browsers_bind.
No parameter schema on file.
Return the full ExportPayload for one chat on the bound browser. Prefer for a complete dump of a known chatId; for turn index/previews use chat_outline, for a bounded slice use chat_read, for offline SQLite reuse use chat_cache_*. Empty chatId = currently open tab. Requires browsers_bind.
No parameter schema on file.
Build a turn index with short previews for a chat (auto-warms the SQLite cache for long threads). Prefer before chat_read to choose start/end indexes; use chats_get for the full payload, chat_cache_read when the cache is already warm. Empty/omitted chatId = open tab. Requires browsers_bind.
No parameter schema on file.
Read a bounded slice of chat turns with a maxChars cap (uses/auto-fills the SQLite cache). Prefer after chat_outline for start/end indexes; use chats_get for the entire chat, chat_cache_read for offline cache-only access. Requires browsers_bind when the cache miss needs a host fetch.
No parameter schema on file.
Export one chat via extension builders and write files under exportDir (returns paths + host gate). Prefer for durable files; use chats_get for in-memory payload, jobs_enqueue kind=export/export_batch for bulk/async queues. Set async:true to return job_id instead of blocking. Captcha is never…
No parameter schema on file.
Create a new chat on the provider UI and seed it with text (no reply wait). Prefer chat_ask when you also need the assistant reply; use chats_send for follow-ups. Supports model|mode|toggles|tools|files|onCaptcha|ensureTab. Captcha is never auto-solved. Requires browsers_bind and allowCreate.
No parameter schema on file.
Send a follow-up message in an existing chat without waiting for the assistant reply. Prefer chat_ask when you need the reply; use chats_create to open a new chat. Supports model|mode|toggles|tools|files|onCaptcha|ensureTab. Captcha is never auto-solved. Requires browsers_bind.
No parameter schema on file.
Send a prompt (new or existing chat) via the bound browser and wait for the assistant reply. Prefer over chats_create+chats_send when you need the reply; use wait:false or jobs_enqueue kind=ask for non-blocking. Supports model|mode|toggles|tools|files|onCaptcha|ensureTab. Captcha is never…
No parameter schema on file.
Fetch a chat from the bound browser and store blocks+FTS5 in the temp SQLite cache. Prefer before repeated chat_cache_read on long threads; use chat_outline/chat_read when you need live outline/slice without an explicit sync. Writes local cache only; requires browsers_bind for the host fetch.
No parameter schema on file.
Read cached chat blocks from local SQLite by turn range, blockId, or FTS query without hitting the browser. Prefer after chat_cache_sync/chat_outline; use chat_read when the cache may be cold and a live fetch is OK. Identify the copy via cacheId or platform+chatId.
No parameter schema on file.
Append turns to an existing local SQLite cache copy and refresh FTS/totals. Prefer for incremental offline updates; use chat_cache_sync to rebuild from the live tab, chat_cache_update_block to edit one turn. Does not write to the provider UI.
No parameter schema on file.
Update one cached block (bumps revision) or resync that turn from the live tab via resyncFromHost. Prefer for surgical cache edits; use chat_cache_append to add turns, chat_cache_sync to refresh the whole chat. Local cache only unless resyncFromHost needs the bound browser.
No parameter schema on file.
Delete a temp SQLite cached chat (metadata, blocks, FTS). Does not delete the provider chat — use chats_delete for that. Identify via cacheId or platform+chatId. Idempotent on missing rows.
No parameter schema on file.
List projects/folders (containers) on a platform via the bound browser. Prefer projects_list (MCP alias containers_list was removed). Use projects_chats for chats inside one project, chats_list for all chats. Wire op remains containers.list. Requires browsers_bind.
No parameter schema on file.
List chats inside one project/container on the bound browser. Prefer over chats_list when scoped to a known containerId; use projects_list first to discover ids. Requires browsers_bind.
No parameter schema on file.
Create a project/folder on the provider UI. Prefer when organizing chats; does not create chats (use chats_create/chat_ask). Requires browsers_bind and allowProjects on the browser.
No parameter schema on file.
Permanently delete a project/folder on the provider. Destructive: requires confirm:true, allowProjects, and allowDestructive on the browser. Prefer projects_move_chat to rehome chats first. Does not delete local SQLite cache rows.
No parameter schema on file.
Move a chat into a project, or pass containerId null to remove it from its project. Prefer over delete when reorganizing. Requires browsers_bind and allowProjects.
No parameter schema on file.
Permanently delete a chat on the provider UI. Destructive and audited: requires confirm:true plus allowDestructive on server and browser. Prefer chats_export first if you need a copy; chat_cache_delete only removes local cache. Captcha is never auto-solved.
No parameter schema on file.
Enqueue a background job on the bound browser lane (ask, export, export_batch, delete, delete_batch) and return job_id immediately. Prefer for long/bulk work; use chat_ask/chats_export synchronously when you can wait. Supports onCaptcha for ask/export/delete kinds. Captcha is never auto-solved.…
No parameter schema on file.
Get status, progress, result, or error for a job_id from jobs_enqueue. Prefer for polling one job; use jobs_list to browse many. Local job store only.
No parameter schema on file.
List background jobs newest-first with optional status/kind/browser filters. Prefer over repeated jobs_status when surveying the queue; use jobs_status for one job_id detail.
No parameter schema on file.
Cancel a queued job immediately, or request cancel for a running job. Prefer when stopping work started via jobs_enqueue/chat_ask wait:false. Does not roll back provider-side side effects already applied.
No parameter schema on file.
Free
$0 forever
- 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 Chat Export 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: —
—Grade not published
No rule-by-rule report on file for this release
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.
This listing ships as a container image. The scanner reads packages on npm and PyPI and public source; an image is neither, so there is no grade here — not a withheld one, an unmeasured one.
Release history
It installs from a container image rather than from a versioned package, so the command cannot be pinned to a release. The history below is for reading; the current release on file is 1.0.8.
1.0.8
13 Nov 2026
latest- MCP platforms: Kimi, Z.ai, MiniMax (JanitorAI stays extension-only). Host-aligned skins. Z.ai create/send reliability fix.
—
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 Chat Export 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.