MCP tools
These are the tools the agentry MCP server exposes to your harness. The agent calls them — you don't type them. But knowing what they do helps you prompt well and understand what the agent is doing on your behalf.
Your harness's tools/list is always the authoritative, up-to-date inventory; this page is the friendly tour.
Sandbox lifecycle
| Tool | What it does |
|---|---|
sandbox_create | Spin up a fresh, isolated sandbox. Returns the sandbox ID and the services already bound to it. |
sandbox_list | List sandboxes, or show details of one (status only — not for choosing where to work). |
sandbox_delete | Tear a sandbox down. |
When sandbox_create returns, it tells the agent which services are already wired in (e.g. a database, an AI key) so the agent uses the real thing instead of stubbing a fake.
Running commands
| Tool | What it does |
|---|---|
command_run | Run a command and wait for it to finish (installs, tests, git, curl). |
command_start | Start a command in the background; returns an ID. |
command_logs | Read new output from a background command. |
command_interrupt | Stop a background command. |
Long-running projects
For dev servers, watchers, schedulers — anything meant to keep running.
| Tool | What it does |
|---|---|
project_create | Scaffold a managed project (Next.js, static HTML, Streamlit, FastAPI, a Python script, or custom). |
project_start | Start (or restart) the project. |
project_stop | Stop it. |
project_list | List projects with status, ports, uptime, and last error. |
project_logs | Read recent logs, with an optional text filter. |
Files
| Tool | What it does |
|---|---|
file_read | Read a file (optionally a line range). |
file_write | Create or overwrite a file (makes parent folders automatically). |
file_list | List files under a folder, optionally recursive and filtered. |
file_grep | Search across many files with a regex. |
file_replace | Replace an exact string in one file. |
file_multi_edit | Apply several edits to one file in a single step. |
Ports
| Tool | What it does |
|---|---|
port_wait | Wait until a port is accepting connections — used after starting a dev server. |
Services & secrets
| Tool | What it does |
|---|---|
service_list | List services that can be bound (Postgres, Redis, MySQL, MongoDB, S3, SMTP, Stripe, OpenAI, Anthropic, and more). |
service_bind | Wire a service into the sandbox; returns the env var names it set. |
secret_set | Set a non-secret env var the app needs (rejects secret-shaped values — use agentry env set for real secrets). |
secret_list | List env var names in the sandbox (never values). |
Checking the app
| Tool | What it does |
|---|---|
app_probe | Make one HTTP request to the running app and report status, latency, content type, and a body snippet. The right way to confirm "is it actually serving?". |
service_probe | Check that a bound service (database, etc.) is reachable from the sandbox. |
deployment_status | Read the status of a sandbox's deployment(s) from the dashboard — status, URL, revision, message. Read-only; it never starts a deploy. |
Python / data
| Tool | What it does |
|---|---|
code_exec | Run Python in a persistent kernel — state carries across calls; charts render inline. Great for data work. |
code_close | Shut down a kernel and free its memory. |
Built-in docs
| Tool | What it does |
|---|---|
docs_read | Read agentry's built-in cheat sheets baked into every sandbox — including the conventions for reading the signed-in user and framework setup tips. |
Next
- CLI reference — the operations you run from a terminal.
- Other MCP clients — wire any MCP-speaking tool to agentry.
- Prompting patterns — prompt so the agent uses these well.