google-cloud-vision
Use this skill when an agent needs Google Cloud Vision API for still-image understanding: labels, object localization, OCR, document text, SafeSearch, image properties, crop hints, web detection, batch annotation, Cloud Storage based pipelines, confidence evaluation, privacy, quotas, cost, and QA. Do not use it for Gemini multimodal reasoning, video analysis, image generation, custom model training, product catalog search design, or human reference/authenticity review.
Profile →npx skills add calesthio/generative-media-skills --skill google-cloud-vision --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Guides an agent to use Google Cloud Vision for still-image understanding, focusing on structured annotations from images such as labels, object bounding boxes, OCR text, SafeSearch, color properties, crop hints, and web-detection signals. It treats Vision as an annotation service rather than a conversational image reasoner.
How it works
- Specifies feature types supported (e.g.,
TEXT_DETECTION,DOCUMENT_TEXT_DETECTION,LABEL_DETECTION,OBJECT_LOCALIZATION,SAFE_SEARCH_DETECTION,IMAGE_PROPERTIES,CROP_HINTS,WEB_DETECTION). - Recommends using
images:annotatefor synchronous requests (up to 16 images) andimages:asyncBatchAnnotatefor offline large batches (up to 2000 images). For PDFs/TIFFs, notes specific limits and output behavior. - Distinguishes when to prefer
TEXT_DETECTIONvsDOCUMENT_TEXT_DETECTIONbased on image type (sparse vs dense text) and instructs to rely onfullTextAnnotationstructure for dense OCR. - Advises privacy considerations, data handling, and region endpoints (us/eu) for OCR features, plus guidance to copy images to Cloud Storage for reliability.
- Provides a feature-by-feature mapping: what output fields to expect (e.g.,
textAnnotations,fullTextAnnotation,localizedObjectAnnotations, etc.). - Includes quotas, pricing notes, and a cost-estimation heuristic: billable_units = number_of_images_or_pages * number_of_paid_features_per_image, with reminders about free tier and regional considerations.
When to use it
Use when the task requires stable, feature-specific annotations from still images (labels, bounding boxes, OCR, SafeSearch, crop suggestions, web signals) and when images reside in or are processed via Cloud Storage or need deterministic fields rather than open-ended reasoning. Do not use for Gemini multimodal reasoning, video analysis, image generation/editing, or custom-model training.
What it can touch
- Tools allowed:
claude-code,codex,copilot,cursor(as declared) - Operations involve synchronous or asynchronous Vision API calls, handling of
images:annotate,images:asyncBatchAnnotate,files:annotate, andfiles:asyncBatchAnnotate, and management of Cloud Storage inputs/outputs according to the workflow guidance.
Caveats
- Rules specify not to promise production SLAs or budgets; emphasizes verifying current Google documentation and regional data handling nuances.
- Notes that external image URLs are not guaranteed to succeed and recommends copying images into Cloud Storage.
- Highlights data usage policies: Vision may not be used to prove authorship or rights clearance; online processing does not persist to disk, while asynchronous batch processing may briefly store data with a short TTL.
- Includes several production heuristics and thresholds that should be rechecked against current quotas, pricing, and feature behavior.
# Google Cloud Vision API for still-image understanding Use Google Cloud Vision when the production task needs structured annotations from still images: general labels, object boxes, text/OCR, dense document text, explicit-content likelihoods, dominant colors, crop suggestions, or web-reference signals. Treat it as a computer-vision annotation service, not as a conversational image reasoner. Verif
What does the google-cloud-vision skill do?
Use this skill when an agent needs Google Cloud Vision API for still-image understanding: labels, object localization, OCR, document text, SafeSearch, image properties, crop hints, web detection, batch annotation, Cloud Storage based pipelines, confidence evaluation, privacy, quotas, cost, and QA. Do not use it for Gemini multimodal reasoning, video analysis, image generation, custom model training, product catalog search design, or human reference/authenticity review.
How do I install it?
Run `npx skills add calesthio/generative-media-skills --skill google-cloud-vision --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From calesthio/generative-media-skills, a repository with 112 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.