drop2run-mcp
OfficialServer Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DROP2RUN_TOKEN | No | Access token. Not needed to start the server or list its tools; every tool call requires it. | |
| DROP2RUN_API_URL | No | API base URL. Defaults to the hosted service. |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| loginA | Signs in through a browser on this machine and stores an access token, which every other tool then uses. Opens the browser here and waits for the person to approve. If it answers that nothing has been approved yet, call it again to keep waiting — that is not a failure. Needs no command line and no token pasted by hand. Where no browser can open on this machine — a container, a remote host — use login_code instead. |
| login_codeA | Signs in where no browser can be opened on this machine — a container, a remote host. The first call returns a short code and a URL to enter it at, which the person opens anywhere; call it again to wait for the approval and store the token. |
| publish_filesA | Publishes one or more files written here — markdown, HTML, CSS, JSON, source code, any text — as a site, and returns its URL. A single page goes at index.html; a single .md, .markdown or .pdf file is a site on its own, served through the documents viewer. Text only: a PDF or an image has to be published from disk with publish_dir. To hold an address before the content exists — the person asks for the site first and the files later — publish a placeholder index.html with |
| publish_dirB | Publishes a folder from disk and returns its URL. The folder needs an index.html at its root, or documents (.md, .pdf, .docx, .xlsx, .epub), which are served through the reader. |
| delete_siteA | Takes a site down permanently — its files, its history and its subdomain. Nothing here undoes it, and the subdomain stays reserved — nobody can use it again. Requires |
| change_subdomainA | Moves a site to a new subdomain. The old one keeps working by redirecting to the new one, and nobody else can take it. Each site gets only a couple of changes, and a change back to an old name counts too, so use this only when the person named the new address. Requires |
| list_sitesA | Lists the sites on this account, newest first, so a publish can go to one of them. Returns up to 100 with the account's total; there are no pages to fetch beyond that. Use get_site for one site's settings and versions. |
| list_foldersA | Lists the folders on this account, each with the path a publish's |
| get_siteA | Reads one site's settings and state — serving mode, password, forms, scheduled takedown, folder, whether it is paused — and its versions, newest first, with the one being served marked. Also says which settings the account's plan does not include. |
| update_siteA | Changes one or more settings of an existing site; anything left out stays as it is. Takes effect on every edge within about a minute, without publishing again. Settings the plan does not include are refused with the reason. Only change what the person asked for — never invent a password or a date. |
| pause_siteA | Takes a site off the air without deleting anything: visitors see that it is paused, and its files, versions and subdomain are kept. resume_site puts it back. |
| resume_siteA | Puts a paused site back on the air. Refused, with the reason, when the plan has no room for another active site, the account is past due, or the site was suspended by Drop2Run. |
| rollback_siteA | Serves an earlier version of a site again, without publishing anything. get_site lists the versions; one whose files were collected cannot be rolled back to. |
| list_peopleA | Lists the people on a site's list — the ones invite_people added — with what each may do ( |
| invite_peopleA | Adds email addresses to a site's list and emails each new one an invitation from the account. Use only addresses the person gave you — never guess, infer or look one up. |
| remove_personA | Takes one address off a site's list. On an invite-only site their access ends within a minute; putting them back is another invitation, with another email. |
| list_commentsA | Reads the comments people left on a site's pages, a thread at a time: the page, the words or element each thread points at, and every comment in it. Open threads by default. Pass |
| reply_commentA | Answers a comment thread on a site as its owner. The reply shows on the page under the owner's name, and the people in the thread are emailed. Publish a fix before replying that it is fixed. The thread stays open unless |
| resolve_commentA | Marks a comment thread on a site resolved, as its owner. It moves from the open threads to the resolved ones of the version the site serves now. Resolve when the person asks; otherwise leave it for the commenter to confirm the fix. reopen_comment undoes it. |
| reopen_commentA | Opens a resolved comment thread on a site again, as its owner. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 20 tools
Each tool targets a distinct resource and action: publishing (publish_files vs publish_dir, split cleanly by text-vs-disk), auth (login vs login_code, split by browser availability), site lifecycle, people, and comments. Overlaps like pause_site/delete_site and resolve_comment/reopen_comment are explicitly differentiated in the descriptions. No two tools appear to do the same thing.
Every tool follows a consistent snake_case verb_noun pattern (publish_files, list_sites, get_site, update_site, pause_site, invite_people, reply_comment, resolve_comment, reopen_comment). The only verb-only names, login/login_code, form a clear pair and don't break readability.
20 tools is slightly on the heavy side but each maps to a real operation across four coherent areas (publishing, auth, site management, people, comments). Nothing looks like filler, though it sits at the upper edge of a comfortable set.
The surface covers the full lifecycle: publish, read, update, pause/resume, rollback, delete, plus people and comment management. Minor gaps exist (e.g. no dedicated site-creation or folder-management tool, and folder creation is only in the dashboard), but core workflows are complete.