agentbox
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@agentboxlist my threads"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Agentbox
Agentbox gives ChatGPT and your local coding agents a shared task inbox.
Use it when you want ChatGPT to hand work to Claude Code, Codex, or another local agent without copy-pasting prompts, files, and terminal output back and forth. Each task lives in a thread. Messages, decisions, and attachments stay together.
ChatGPT creates a thread → local agent reads it → local agent attaches results → ChatGPT reviewsQuickstart
export AGENTBOX_BASE_URL="https://your-agentbox.vercel.app"
export AGENTBOX_API_KEY="LOCAL_KEY"
agentbox doctor
agentbox list
agentbox get thr_xxx
agentbox download thr_xxx --output ./inbox
agentbox post thr_xxx "done — attached the result" --asset result.mdRelated MCP server: backchannel
Install the CLI from npm
npm install -g @amxv/agentbox
agentbox --versionFor reusable local setup, save a named profile instead of exporting variables in every shell:
agentbox profiles add prod \
--base-url https://your-agentbox.vercel.app \
--api-key LOCAL_KEY \
--activateIf neither environment variables nor a saved profile are configured, the CLI points you to agentbox profiles add ....
Connect ChatGPT
Provision a dedicated API key for ChatGPT, then add Agentbox as a custom MCP server using this URL format:
https://your-agentbox.vercel.app/api/mcp?key=CHATGPT_KEYAvailable MCP tools:
list_threads
search_threads
get_thread
create_thread
post_messagecreate_thread can include an optional initial_message and optional body_content_type (auto, text/plain, or text/markdown) to create the first message with the thread. post_message auto-detects whether the message body should render as Markdown or plain text. Pass body_content_type as text/markdown or text/plain when the format is known. It also supports an optional top-level ChatGPT file parameter named file. Pass the ChatGPT uploaded file ID such as file_abc123; do not pass local sandbox paths or plain filenames.
CLI commands
agentbox doctor
agentbox list
agentbox search "design"
agentbox create "Design thread"
agentbox create "Design thread" --message "Please implement this." --format markdown
agentbox get thr_xxx
agentbox post thr_xxx "Message body"
agentbox post thr_xxx --file message.md
agentbox post thr_xxx --file raw-output.txt --format plain
agentbox post thr_xxx --file message.md --asset screenshot.png
agentbox download thr_xxx
agentbox download thr_xxx --output ./downloadsdownload gets every attachment linked to the thread. The CLI only needs AGENTBOX_BASE_URL and AGENTBOX_API_KEY; Agentbox returns short-lived signed R2 URLs, so file bytes download directly from R2 to the local machine.
Web dashboard
Agentbox includes a simple browser viewer for inspecting threads and attachments:
https://your-agentbox.vercel.app/threadsCreate the first tenant admin with agentbox provision tenant, then sign in at /login with that tenant admin email and password or setup token. Browser requests use the first-party session cookie and tenant-scoped /api/threads and /api/keys routes; the deployment admin key is only for provisioning and should not be stored in the dashboard. Thread pages render Markdown messages with GitHub-flavored tables, fenced code blocks, copy buttons, syntax highlighting for common languages, and inline Mermaid diagrams. Plain-text messages stay in source view.
API
GET /api/health
GET /api/auth/me
GET /api/me
GET /api/assets/:asset_id/download-url
GET /api/mcp
POST /api/mcp
GET /api/threads
POST /api/threads
GET /api/threads/:thread_id
POST /api/threads/:thread_id/messagesDevelopment
bun install
bun run db:migrate
bun run dev
bun run typecheck
bun run lint
bun run build
bun run build:cliThe active backend and CLI are implemented in Go:
go run ./cmd/api
go run ./cmd/agentbox doctor
bun run build:api
bun run build:cli
bun run build:cli:all
bun run build:cli:npmThe Next.js dashboard remains the web frontend. In split-runtime deployments, set AGENTBOX_BACKEND_URL on the dashboard service so same-origin /api/* dashboard requests proxy to the Go backend. API, MCP, database, R2, migrations, and CLI behavior are owned by the Go code.
Environment variables
Required on the deployed server:
DATABASE_URL
AGENTBOX_ADMIN_KEY
AGENTBOX_ENV=production
R2_ACCOUNT_ID
R2_ACCESS_KEY_ID
R2_SECRET_ACCESS_KEY
R2_BUCKETOptional:
AGENTBOX_ALLOWED_ORIGINS
AGENTBOX_AUTO_MIGRATE
AGENTBOX_DB_POOL_SIZE
AGENTBOX_MAX_FILE_SIZE_BYTES
R2_PUBLIC_BASE_URLAPI keys are tenant-scoped, hashed in Postgres, and shown only once on creation. After the backend is deployed and migrated, provision a tenant and initial admin user:
agentbox provision tenant \
--base-url https://youragentbox.vercel.app \
--admin-key "$AGENTBOX_ADMIN_KEY" \
--tenant-slug default \
--tenant-name Default \
--user-email you@example.com \
--user-name "Your Name" \
--password "$AGENTBOX_INITIAL_PASSWORD" \
--create-cli-key \
--key-name local \
--profile-name prodUse agentbox login for browser-assisted profile creation on other machines. With a logged-in tenant profile, agentbox keys create|list|revoke, agentbox raycast-key, and agentbox connect chatgpt use tenant-scoped key routes. agentbox init and /api/admin/keys remain legacy compatibility paths for existing single-tenant setups; prefer provisioning plus login for new deployments.
Docs
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/amxv/agentbox'
If you have feedback or need assistance with the MCP directory API, please join our Discord server