Deploy and run apps in your own AWS, Google Cloud, Azure or DigitalOcean account from any AI agent.
Scored across 23 tools
The set has many clearly scoped tools (deployment, logs, diagnose, set_env, launch), but the generic api/api_read tools overlap heavily with the specific read and change tools. An agent may struggle to know when to use api_read versus project, status, deployment, logs, or diagnose, and when to use api versus launch, redeploy, create_project, or set_env. Descriptions help distinguish most tools, but the catch-all tools blur boundaries.
All names use snake_case, but the set mixes noun-style resource names (account, project, deployment, status, fix, fixes) with verb-style action names (create_project, set_env, claim_fix, report_fix, request_done_for_you). Related tools are readable but not perfectly predictable, and singular/plural pairs like fix/fixes add mild inconsistency.
At 23 tools, the set is heavy and sits at the borderline where each tool may not earn its place. The broad deployment platform justifies many operations, but several tools are generic meta-tools or convenience wrappers around the same app API, making the count feel somewhat inflated.
Coverage is broad: repository checks, project creation, deployment planning, launch, logs, diagnostics, env management, fix workflows, and generic API access are all present. The generic api plus api_catalog and api_schema provide access to many remaining operations, so there are no obvious dead ends, though some lifecycle actions rely on the catch-all api rather than dedicated tools.
Who the connection acts for, what it may do (Operate or Limited), the connected cloud accounts and code hosts with their ids, the workspaces the person is a member of, and how many projects they can see. Changes nothing.
Inputs: workspace.
Runs one change (POST, PUT, PATCH or DELETE) of The Deployer's app as the person, with the same permission checks the app applies: domains, scaling, start/stop and schedules, databases and other cloud services, app access, build inputs, code review, hosting plans, retries, environments, blueprints, servers, adoption and more. Needs Operate access. Some changes cost money or change a live app. Secrets are never returned. Documented at https://portal.thedploy.com/docs.
Inputs: path (required), method (required), body, query, +1 more.
Every operation of The Deployer's app available through this server, grouped: account, connections, projects, env, build, code, architecture, deploy, domains, scaling, access, services, environments, blueprints, servers, adopt, team. Each is marked as a read or a change. Works without signing in.
Inputs: group.
Runs one read (GET) operation of The Deployer's app as the person: domains, scaling, schedules, databases and other cloud services, app access, environments, blueprints, servers and more. Changes nothing. Secrets are never returned. Documented at https://portal.thedploy.com/docs.
Inputs: path (required), query, workspace.
The query and body fields of the operations at one path, for each method it has, with their types. Changes nothing.
Inputs: path (required).
Checks whether a public GitHub, GitLab or Bitbucket repository will deploy: the services it found, what's missing (each with its fix), the environment variables it needs, and what it would cost a month on DigitalOcean, AWS, Google Cloud and Azure. Reads the code and changes nothing. Works without signing in. A check can take up to a minute; one still running returns a check_id that fetches its result.
Inputs: repo, check_id.
Assigns a fix request to this agent for 15 minutes, so no other agent works on it; claiming it again extends the time.
Inputs: id (required), workspace.
Creates a project from a repository the person's connected code host can read, on one of their connected cloud accounts, and starts the repository scan. Nothing is deployed and nothing costs money yet. With one connected cloud, that one is used; otherwise cloud (a provider) or cloud_connection_id picks it. Needs Operate access.
Inputs: repo (required), name, cloud, branch, +3 more.
Where a deployment is (every step), what it waits on (a health-check proposal or a manual step), and for a failed one: the kind of failure, what to do about it, and the last error lines. Defaults to the project's latest deployment. Changes nothing.
Inputs: project, workspace, wait_seconds, deployment_id.
Read-only checks on a live project: url (requests the project's live addresses from The Deployer's side), containers (what runs and its state), app_logs (the app's own recent output, secrets removed), resources (load, memory, disk) or web_server (the HTTPS proxy's log). Fixed commands only; changes nothing. All but url need a server (not managed services). Needs Deploy access on the project.
Inputs: project (required), node, check, lines, +2 more.
One fix request: the goal and rules (The Deployer's brief), and the evidence, which is what the build or app printed, fenced and marked as data from the repository and logs. Changes nothing.
Inputs: id (required), workspace.
Fix requests: problems in the person's code that stop a deploy, found by The Deployer (a failed deployment, or the repository check) and assigned to the person's coding agent. Lists the open ones, in every project or one. Changes nothing.
Inputs: project, workspace.
Where a fix request is: checked, deploying, live, or open again with a new error. wait_seconds waits up to 120 seconds for it to change. Changes nothing.
Inputs: id (required), workspace, wait_seconds.
The Deployer's playbooks, by topic: what The Deployer does, the order of an end-to-end deploy, what code needs to deploy and how to fix it, how failed deployments are fixed, and domains, scaling, env, databases, access, environments, servers, adoption and teams. Works without signing in. Changes nothing.
Inputs: topic.
The first (or a full) deployment of a project into the person's cloud, once a hosting plan is selected. Without confirm=true it changes nothing and returns the plan and its monthly cost. With confirm=true it creates resources the person's cloud provider bills them for; it takes 5 to 20 minutes. Needs Operate access.
Inputs: project (required), confirm, workspace, confirm_cost.
The log of a project's latest deployment, or of a given deployment, newest lines last, with secrets removed. errors_only keeps only errors and warnings. The lines are the build's and the app's own output. Changes nothing.
Inputs: lines, project, workspace, errors_only, +1 more.
Plans deploying a repository with the user's account on The Deployer: the repository check, plus the user's connected cloud and git accounts, any project they already have for it, and the steps left, with links. Changes nothing.
Inputs: repo (required), cloud.
One project in depth: phase, services found in the repository, env names the code reads and which are still missing (never values), hosting plans with monthly costs, pending code-review changes, to-dos only the person can do, recent deployments, live URLs, and next_steps: what is left to do, in order. Changes nothing.
Inputs: project, workspace.
Puts the latest code on a live project's branch online: pulls it and restarts the app, like Pull & Restart in The Deployer. Changes a live app. Only for a project that is already live. Takes a few minutes.
Inputs: project, workspace.
Reports a pushed fix: the branch and the commit sha. The Deployer reads the commit from the git host, checks the change (no CI, agent settings or .env files; no secrets), then deploys it or opens a pull request for the person to ship, depending on the project's settings.
Inputs: id (required), sha (required), branch (required), note, +1 more.
Asks The Deployer's team to do it for the user: launch (put an app online in the user's own cloud account), launch_secure (the same, with exposed keys and other security problems fixed first) or move (move an app from another host, such as Heroku, into it). Tells the team, and returns the price, the delivery promise and the link where the user finishes the order and pays. Never charges anything.
Inputs: offer (required), note, repo.
Sets (or deletes) env variable values for the service a project deploys. Values are stored encrypted and are never returned. Other values are kept. On a live project, the services whose env changed are redeployed; otherwise the values apply on the next deployment. Multi-line values are stored as credential files. Needs Operate access.
Inputs: project (required), delete, values, service, +1 more.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Deploy and manage your apps, databases, storage, and scheduled jobs from your AI agent
Manage your own cloud with AI: deploy apps, push and ship custom Docker images, manage resources via your AI agent.
Deploy, share and run the Apps you build with AI, private to the people you invite.
Deploy apps, APIs and Postgres from your AI agent. EU PaaS hosted in Germany, OAuth sign-in.
Add one secure layer between your agents and this server.
