Skip to content

features-by-page.md

Latest commit

 

History

History
118 lines (96 loc) · 6.53 KB

features-by-page.md

File metadata and controls

118 lines (96 loc) · 6.53 KB
Raw

Roboflow: Feature Lookup by Intent

Source-of-truth note: This page ships with the Roboflow plugin. If your client has the plugin loaded, prefer the local skill (roboflow:product-navigation) over fetching roboflow://skills/product-navigation/features-by-page via ReadMcpResourceTool — the MCP resources are a fallback for non-plugin clients and may lag the source repo.

"I want to do X" -> go here / use this tool.

Base URL: https://app.roboflow.com

Upload Data

Intent Web URL Alternatives
Upload images/videos /{ws}/{proj}/upload Python SDK: project.upload(path), MCP: image_upload + image_upload_status
Import from S3/GCS/Azure /{ws}/{proj}/upload -> Cloud Import tab Python SDK with cloud URLs
Import from Universe /{ws}/{proj}/upload -> Universe tab MCP: universe_search then fork
Upload pre-annotated data /{ws}/{proj}/upload (drag folder with annotations) Python SDK: project.upload(path) auto-detects annotations

Annotate

Intent Web URL Alternatives
Draw bounding boxes /{ws}/{proj}/annotate --
Draw polygons/masks /{ws}/{proj}/annotate (select polygon/mask tool) --
Annotate keypoints /{ws}/{proj}/annotate (keypoint project) --
Use AI auto-label /{ws}/{proj}/annotate/batch/{batchId}/autoLabel MCP: annotations_save with model predictions
Use Label Assist (model-assisted) /{ws}/{proj}/annotate -> toggle Label Assist --
Use Smart Polygon (SAM) /{ws}/{proj}/annotate -> Smart Polygon tool --
Assign labeling to team /{ws}/{proj}/annotate/batch/{batchId}/createJob MCP: annotation_jobs_create
Review annotation batches /{ws}/{proj}/annotate MCP: annotation_batches_list, annotation_batches_get
Check annotator productivity /{ws}/settings/insights --

Manage Dataset

Intent Web URL Alternatives
Search images /{ws}/{proj}/images -> search bar MCP: images_search
Filter by class/tag/split /{ws}/{proj}/images -> filter panel --
Add tags to images /{ws}/{proj}/images -> select images -> Tag --
Manage classes /{ws}/{proj}/settings --
Delete images /{ws}/{proj}/images -> select -> Delete --
Check class balance /{ws}/{proj}/health --
Merge projects /{ws}/merge --
Make project public/private /{ws}/{proj}/sharing --

Generate Dataset Version

Intent Web URL Alternatives
Create version with preprocessing /{ws}/{proj}/generate MCP: versions_generate
Add augmentations /{ws}/{proj}/generate -> Augmentation step MCP: versions_generate (augmentation params)
Set train/test split /{ws}/{proj}/generate -> first step MCP: versions_generate
View version details /{ws}/{proj}/{version} MCP: versions_get
Export/download version /{ws}/{proj}/{version}/export MCP: versions_export, Python SDK: version.download(format)

Train Models

Intent Web URL Alternatives
Train a model /{ws}/{proj}/train MCP: trainings_create
Choose model architecture /{ws}/{proj}/train -> architecture step MCP: trainings_create (model param)
Train from checkpoint /{ws}/{proj}/train -> checkpoint step MCP: trainings_create
Train specific version /{ws}/{proj}/{version}/train MCP: trainings_create
Check training status /{ws}/{proj}/{version} (shows progress bar) MCP: models_get_training_status
View training results (mAP, etc.) /{ws}/{proj}/{version}/train/results MCP: models_get
Cancel training /{ws}/{proj}/{version} -> Cancel button --
Use Roboflow Instant (auto-train) Triggers on batch approval --

Deploy & Inference

Intent Web URL Alternatives
Get API snippet for model /{ws}/{proj}/deploy --
Run inference on image /{ws}/{proj}/deploy -> Try tab MCP: models_infer
Set up dedicated deployment /{ws}/deployments?tab=dedicated API
View edge devices /{ws}/deployment-manager/devices --
Batch processing /{ws}/deployments?tab=batch --
Monitor model performance /{ws}/vision-events --
Upload custom model weights /{ws}/{proj}/deploy -> Upload Python SDK: version.deploy(model_type, path)

Workflows

Intent Web URL Alternatives
Create workflow /{ws}/workflows -> Create --
Edit workflow (visual) /{ws}/workflows/{id} --
Edit workflow JSON /{ws}/workflows/{id} -> JSON tab MCP: workflow_specs_validate
Test workflow /{ws}/workflows/{id} -> Preview/Test MCP: workflows_run, workflow_specs_run
Deploy workflow /{ws}/workflows/{id} -> Deploy Inference SDK, REST API
List available blocks /{ws}/workflows/{id} -> block palette MCP: workflow_blocks_list

Universe (Pretrained Models & Datasets)

Intent Web URL Alternatives
Search for datasets universe.roboflow.com -> search MCP: universe_search
Search for models universe.roboflow.com -> search (filter: has model) MCP: universe_search
Fork dataset to workspace universe.roboflow.com/{user}/{proj} -> Fork --
Try a pretrained model universe.roboflow.com/{user}/{proj}/model MCP: models_infer with universe model ID
Download dataset universe.roboflow.com/{user}/{proj} -> Download Python SDK: rf.universe(user, proj).version(v).download(fmt)

Team & Workspace Management

Intent Web URL Alternatives
Invite team member /{ws}/settings/members -> Invite --
Change member role /{ws}/settings/members -> click member --
Create project folders /{ws} -> Folders section --
Set folder permissions /{ws} -> Folder -> Settings --
View/change plan /{ws}/settings/plan --
Check credit usage /{ws}/settings/usage --
Purchase credits /{ws}/settings/plan -> Buy Credits --
Update payment method /{ws}/settings/plan -> Payment tab --
Get workspace API key /{ws}/settings/api --
Get personal API key /settings/account -> API Keys --
Configure SSO /{ws}/settings/plan (Enterprise) --
View audit logs /{ws}/settings/audit-logs (Enterprise) --

For plans, credits, and cost estimation, see roboflow://skills/plans-and-pricing/SKILL.