Get a Roboflow workspace's metadata and its projects.

Pass workspace to override the default set in ROBOFLOW_WORKSPACE. Set dry_run=True to preview the HTTP request without calling the API.

List all projects in a Roboflow workspace.

Lighter than roboflow_get_workspace when you only need the project list. Pass workspace to override ROBOFLOW_WORKSPACE. Set dry_run=True to preview.

Search images in a Roboflow project.

Filter by tag (e.g. "sku-42"), class_name, or a semantic prompt. Paginated via limit (max 250) and offset. workspace falls back to ROBOFLOW_WORKSPACE. Set dry_run=True to preview the HTTP request without calling.

Add one or more tags to an image. Additive; not destructive.

Remove tags from an image.

Destructive: requires confirm='yes' and a server mode of curate or full (see :envvar:ROBOFLOW_MCP_MODE).

Replace an image's tags with the given list.

Destructive: requires confirm='yes' and a server mode of curate or full (see :envvar:ROBOFLOW_MCP_MODE).

Upload a single image to a Roboflow project.

source is a discriminated union (exactly one mode):

• {"kind": "url", "url": "https://..."}

• {"kind": "path", "path": "/abs/path/img.jpg"} (local path; must live under ROBOFLOW_MCP_UPLOAD_ROOTS)

• {"kind": "base64", "data": "...", "filename": "img.jpg"}

Every mode runs through URL / path / image safety guards before the upload. split is "train" / "valid" / "test".

Upload many images concurrently.

sources is a list of :class:ImageSource dicts. Failed uploads don't abort the batch; the response reports per-image outcomes in results and errors.

Delete an image. Destructive: requires confirm='yes' and ROBOFLOW_MCP_MODE=curate or full.

Get the metadata of a single image.

List upload batches for a project.

Attach an annotation to an image.

Supported formats (annotation_format literal):

• "coco"

• "yolo" (YOLOv5/v8/v11-compatible .txt)

• "pascal_voc" (XML)

• "createml" (Apple CreateML JSON)

• "roboflow_json"

Pass annotation as a string (raw file contents) or a dict (Python object; we serialize to JSON). Optional labelmap is a class-id to name map Roboflow uses for YOLO formats.

Read project metadata (classes, image count, splits).

List a project's annotation jobs (Annotate-tab batches).

Each job reports labeling / review progress; ready_to_add marks jobs sitting in review with every image approved, i.e. eligible for roboflow_add_reviewed_to_dataset. Optionally filter by status (e.g. "assigned", "review", "complete").

Add every fully-reviewed annotation job to the dataset.

Moves images from jobs in review with all images approved into the dataset, split randomly at the given global ratios (default 70/20/10). Restrict to specific jobs via job_ids.

Destructive: requires confirm='yes' and a server mode of curate or full. Also requires ROBOFLOW_SESSION_COOKIE (an app.roboflow.com browser-session cookie) because Roboflow has no public API for this action.

List dataset versions under a project.

Read one version's full metadata.

Kick off an async version generation (quota-heavy).

Destructive-of-quota: requires confirm='yes' and ROBOFLOW_MCP_MODE=curate or full. After this returns, poll roboflow_get_version_generation_status until the new version reports status="ready".

Poll the async generation state of a version.

Request an export for a trained/generated version.

Returns a signed download URL when ready. Does NOT stream bytes — pair with roboflow_download_export if you want the zip on disk.

Delete a dataset version. Destructive: confirm='yes' required.

Stream an exported dataset zip to disk.

Writes under ROBOFLOW_MCP_EXPORT_CACHE_DIR (default ~/.cache/roboflow-mcp). dest_dir, when given, must be a subdirectory of the cache root.

Destructive-to-fs: requires confirm='yes' and ROBOFLOW_MCP_MODE=curate or full.