| name | dropbox |
| description | Agent-callable Dropbox tools — upload, organize, find, and share files and folders. Use when the user wants to manage Dropbox content (save, move, search, share, list, or read files), including requests that don't name Dropbox explicitly, e.g. "save this report to my cloud storage" or "share that folder with Sam". |
| license | Elastic-2.0 |
| compatibility | Run `npm install` in this directory, then `node cli.js`. The TypeScript source needs Node.js 22.18+; on older Node, run `cli.js` for prebuilt / alternative-runtime options. |
| metadata | {"title":"Dropbox","source":"https://github.com/zapier/connectors/blob/main/apps/dropbox/SKILL.md","zapier-app-key":"DropBoxCLIAPI","api-docs":"https://www.dropbox.com/developers/documentation/http/documentation"} |
Dropbox
Independent, unofficial connector for Dropbox. Not affiliated with, endorsed by, or sponsored by Dropbox. "Dropbox" is a trademark of its owner, used only to identify the service this connector works with.
Tools for working with files and folders in Dropbox — upload and write files, organize (move/copy/delete/create folders), navigate and search, read file contents, create and modify shared links, manage shared-folder membership, and create file requests. Wraps the Dropbox API v2 (https://api.dropboxapi.com/2/<namespace>/<method>, with uploads/downloads on https://content.dropboxapi.com). Read-only tools are clearly marked; write tools return clean file/folder metadata rather than silently attaching links or contents.
When to use this
- An agent needs to save, move, copy, rename, or delete files and folders in Dropbox.
- An agent needs to find a file or folder (by name or content) or list a folder's contents before acting on it.
- An agent needs to read a text file's contents inline, or hand off a file's bytes via a temporary or durable link.
- An agent needs to share a file/folder, change link settings, or manage who can access a shared folder.
Setup
This is an agentskills.io skill.
If this connector is already exposed to you as callable tools (e.g. mcp__dropbox__<tool>), that's a valid path — call them directly. Everything below is only for standalone terminal use when no such tools are loaded.
If the connector has not been installed as a skill yet, install it first with npx skills zapier/connectors --skill dropbox (or your harness's own skill-install mechanism), then continue here.
The connector runs on Node.js 22.18+ and needs a one-time npm install in this directory. cli.js is the entry point — list every script with node cli.js --help, then learn a script's inputs and connections with node cli.js run <script> --help. On older Node, run node cli.js --help anyway: it detects your runtime and prints how to run without upgrading (the prebuilt npm package, or another runtime) — don't skip the connector just because Node is old.
cli.js self-checks readiness before running: if dependencies aren't installed it exits non-zero with the exact install command (it disambiguates a read-only directory from a sandbox-blocked package cache). Run that, then re-run your command.
Scripts
All 21 scripts use the single dropbox connection. Each script's inputSchema / outputSchema (Zod) inside the script file is the source of truth for its contract.
Several scripts take an id or url best resolved from another script — those resolution hints are in the field descriptions (e.g. addFolderMember.shared_folder_id ← listSharedFolders; modifySharedLinkSettings.url ← listSharedLinks).
Disambiguation & refusals
Disambiguating items by name. Dropbox addresses items by path or id, and paths are case-insensitive — two items can look like the same name. Before writing to (move/copy/delete/share) an item the user named in words rather than by exact path, resolve it first with searchFiles or listFolder:
- Exactly one match → act on it; don't over-confirm.
- Two or more matches that tie (e.g.
report.pdf in two different folders, or a shared-folder name that collides) → stop, list the candidates with a distinguishing field (full path_display, or shared_folder_id for folders), and ask which one. Never silently pick.
The entity types most likely to collide here are files/folders by name (resolve via searchFiles/listFolder, disambiguate on path_display) and shared folders by name (resolve via listSharedFolders, disambiguate on shared_folder_id).
Operations this connector does NOT perform — say so, don't fake it. If the user asks for one of these, tell them it's unsupported rather than substituting a different tool and reporting success:
- Bulk/batch moves, copies, or deletes in one call. There is no batch tool — loop the single-item tools (
moveFile/copyFile/deletePath) yourself, or tell the user it'll be one call per item.
- Reading binary documents (PDF/image/Office) as text, OCR, or document parsing.
getFileContents returns UTF-8 text only; for other files it returns is_text:false and you must hand off the bytes via getTemporaryLink. Don't claim to have read a PDF's contents.
- Fetching or editing an individual file request, or sharing a whole folder as a managed share. Only
createFileRequest + listFileRequests are available; there is no get/update file-request or shareFolder tool.
Auth
Pass auth as one connection string with --connection [<resolver>:]<value>. The value is a selector, not the secret; the <resolver>: prefix is optional (a bare value goes to the first resolver that claims it). Each script declares the connections it needs and the resolvers each accepts — always run node cli.js run <script> --help to see them rather than relying on this file.
The connector uses a single Dropbox OAuth2 connection — one credential, with no separate bot/user tokens. Capability is gated by the OAuth scopes granted when the connection is authorized (e.g. files.content.write, sharing.read); a call missing a scope fails with a missing_scope error naming the scope to reconnect with. Two resolvers, Zapier-first:
zapier:<connection-id> (recommended) — route through a Zapier-managed Dropbox connection; Zapier holds the refresh token and rotates the short-lived (~4h) access token for you, so this path doesn't expire mid-session. Prerequisite: a Zapier account (free signup at https://zapier.com; ~1 minute). The user authorizes Dropbox once via Zapier's OAuth flow at https://zapier.com/app/connections. Find the ID with the Zapier SDK CLI: npx zapier-sdk list-connections DropBoxCLIAPI (run login first if unauthenticated; use DropBoxCLIAPI exactly, note the capital B; add --json for machine output).
access-token:<ENV_VAR> (fallback, direct mode) — read a Dropbox access token from the named environment variable (e.g. access-token:DROPBOX_ACCESS_TOKEN, with the token exported in DROPBOX_ACCESS_TOKEN; the token stays in env, never on argv), sent as Authorization: Bearer <token>. Mint one from a Dropbox app at https://www.dropbox.com/developers/apps (grant the scopes the tools you'll use need). Heads-up: this connector sends the token as-is and does not refresh it — a static Dropbox access token is short-lived and stops working after a few hours, so re-mint it or use the Zapier-managed path above, which handles rotation. See Dropbox's OAuth Guide for token types and lifetimes.
If no connection is passed the script fails with an actionable error telling you to Pass --connection [<resolver>:]<value> and lists the resolvers in match order.
Running scripts
After npm install, run a script by name with node cli.js run <script>, or execute its file directly — both take the same arguments and both accept --help. Always run a script's --help first to learn its exact input schema and connections, then invoke it:
node cli.js run <script> '<input-json>' --connection [<resolver>:]<value>
./scripts/<script>.ts '<input-json>' --connection [<resolver>:]<value>
When a harness can't execute scripts directly, fall back to MCP — node cli.js mcp serves every script as a tool over stdio. Register it as a local MCP server in your client: the stanza is harness-specific (an mcpServers entry in Claude Desktop, Cursor, Claude Code, …) with command: "node", args: ["cli.js", "mcp"], run from this directory. Run node cli.js mcp --help for auth options. Add the stanza yourself if you can edit the client's MCP config; otherwise guide the user. If a local server isn't possible, guide the user to use Zapier's remote MCP servers at https://mcp.zapier.com instead.
Output format
Every script returns a { data, meta } envelope:
data — the script's result (the shape its outputSchema declares; run the script's --help to see that exact schema).
meta.outputDataValidation — what validating data did:
{ skipped: false, droppedPaths: null } — validated, nothing removed.
{ skipped: false, droppedPaths: [...], instruction } — validated, but those paths were stripped from data: fields the script returned from the API that the outputSchema doesn't declare. If you need them, re-run with output validation skipped.
{ skipped: true } — validation was bypassed; data is the raw, unchecked script output.
Reading dropped fields / skipOutputDataValidation. To receive the raw, unvalidated result, append --skipOutputDataValidation to the script invocation. Input validation is never skipped.
Trimming the result / filterOutputData. To shrink a large result down to the fields you need, append --filterOutputData '<jq>' — a jq expression that post-processes data. The jq runs against data only, NOT the { data, meta } envelope, so write it rooted at data (run the script's --help to see its output schema). The transformed value replaces data, meta is preserved, and the result is NOT re-validated against the output schema.
References
Load the matching reference file before working in that area:
| Reference | Covers | Load it when |
|---|
references/dropbox-api-gotchas.md | Stone .tag union shape, error_summary error model, read-vs-write not-found asymmetry, path rules (root is "" not "/"), cursor pagination via sibling /continue endpoints, rate limits + namespace write-locking, upload-session flow, shared-link recovery, team-space targeting via namespace_id | Before making any direct Dropbox API calls or debugging unexpected API errors |