# Pictograph SDK > Agent-native computer-vision annotation platform. Python SDK wraps a REST > API exposing dataset upload, SAM3 auto-annotation, model training, and > exports. The SDK ships a tool registry consumable by Claude / OpenAI / > any agent stack via a JSON-schema endpoint. ## Documentation ### Getting Started - [Introduction](https://pictograph.io/docs/index.md): Pictograph is an agent-native computer-vision annotation platform. The SDK ships a Python client, a Claude Skill, an agent toolkit (Claude + OpenAI), and a CLI — all wired to one REST API. - [Installation](https://pictograph.io/docs/installation.md): Install the Pictograph SDK and its optional extras (CLI, agents, cache, telemetry). - [Quick start](https://pictograph.io/docs/quick-start.md): Install the Pictograph SDK, get an API key, and run your first end-to-end pipeline in five minutes. - [Authentication](https://pictograph.io/docs/authentication.md): API key format, resolution order, and the role-based permission model. ### Agents - [Agents — overview](https://pictograph.io/docs/agents.md): Pictograph ships a Claude + OpenAI agent toolkit, a bundled Claude Skill, and a JSON-schema endpoint for dynamic-discovery agents (Vercel AI SDK, LangChain, etc.). - [Claude integration](https://pictograph.io/docs/agents/claude.md): Two integration paths — raw Anthropic tool dicts (zero extra deps) or the Claude Agent SDK with `@tool` decorators. Both backed by the same registry. - [OpenAI integration](https://pictograph.io/docs/agents/openai.md): Two integration paths — raw OpenAI function-tool dicts or the openai-agents SDK with `@function_tool` decorators. - [Bundled Claude Skill (`pictograph-cv`)](https://pictograph.io/docs/agents/skill.md): A wheel-bundled Claude Skill that gives Claude Code / claude.ai workflow recipes, reference docs, and drop-in Python wrappers for common Pictograph operations. - [Dynamic discovery (Vercel / LangChain / etc.)](https://pictograph.io/docs/agents/dynamic-discovery.md): Use `GET /api/v1/developer/tools.json` to wire Pictograph into any agent framework that speaks JSON Schema — no Pictograph-specific adapter required. - [Agent cookbook](https://pictograph.io/docs/agents/cookbook.md): Recipe-style examples — credit-aware training, V7 import + re-annotate, batch SAM3 with progress, multi-class export pipelines. ### API Reference - [API reference — overview](https://pictograph.io/docs/api-reference.md): Index of every Pictograph SDK resource. Each page mirrors a Python `client..()` 1:1 with the underlying REST endpoint. - [Datasets](https://pictograph.io/docs/api-reference/datasets.md): List, fetch, create, delete projects and bulk-download their images + annotations. - [Images](https://pictograph.io/docs/api-reference/images.md): Upload, fetch, download, and delete individual images. - [Annotations](https://pictograph.io/docs/api-reference/annotations.md): Read, save, and delete annotations on individual images. Save is a full overwrite — pass the complete list every time. - [Auto-annotate](https://pictograph.io/docs/api-reference/auto-annotate.md): SAM3 single-prompt (point, box, text) and async batch endpoints for AI-generated annotations. - [Training](https://pictograph.io/docs/api-reference/training.md): Spawn, monitor, and cancel model training runs on Modal GPUs (YOLOX, Detectron2, SM-PyTorch, classification, RF-DETR). - [Models](https://pictograph.io/docs/api-reference/models.md): List trained CV models in your organization and download their ONNX weights. - [Credits](https://pictograph.io/docs/api-reference/credits.md): Balance, ledger history, and pre-flight cost estimation. Gate paid operations to avoid mid-run PaymentRequiredError. - [Exports](https://pictograph.io/docs/api-reference/exports.md): Build and download dataset exports in COCO, YOLO, CVAT, Pascal VOC, LabelMe, CSV, or canonical Pictograph JSON. - [API keys](https://pictograph.io/docs/api-reference/api-keys.md): Programmatic API key management. Requires admin or owner role on the calling key. - [Batch](https://pictograph.io/docs/api-reference/batch.md): Bulk move, copy, delete, and update across many images in one round-trip. - [Search](https://pictograph.io/docs/api-reference/search.md): Find images by SigLIP2 cosine similarity to a reference, or by automatic content tags (objects / scenes / attributes). - [Connectors](https://pictograph.io/docs/api-reference/connectors.md): Import datasets from V7 (Darwin) and Roboflow into Pictograph. Annotations are converted to canonical Pictograph JSON automatically. - [Video](https://pictograph.io/docs/api-reference/video.md): Upload videos, probe metadata, and extract frames as images into a dataset. - [Organizations](https://pictograph.io/docs/api-reference/organizations.md): Read the active organization, manage members, and send / revoke invites. Admin or owner role required for mutations. - [Projects](https://pictograph.io/docs/api-reference/projects.md): Project (dataset) write operations — create, update config (classes / annotation types), delete. - [Agent tool registry](https://pictograph.io/docs/api-reference/tools.md): Fetch the JSON Schema array of every agent-callable tool. The same registry the Python SDK exposes via `Toolkit.as_json_schema()`. ### Reference - [Error handling](https://pictograph.io/docs/error-handling.md): The exception hierarchy, when each error fires, and how to retry safely. - [Rate limits](https://pictograph.io/docs/rate-limits.md): Per-tier request budgets, the sliding window algorithm, and how the SDK handles 429s. - [Pictograph annotation format](https://pictograph.io/docs/annotation-format.md): The canonical Pictograph JSON schema for bbox, polygon, polyline, and keypoint annotations. Class labels go in `name` (not `class`); polygons use multi-ring `paths`. - [CLI reference](https://pictograph.io/docs/cli.md): Every `pictograph` subcommand, mirroring the SDK 1:1. Install via `pip install 'pictograph[cli]'`. ## External resources - [Bundled `tools.json`](https://api.pictograph.io/api/v1/developer/tools.json): Agent tool registry as JSON Schema (requires API key). - [GitHub](https://github.com/pictograph-labs/pictograph-sdk): SDK source. - [PyPI](https://pypi.org/project/pictograph/): `pip install pictograph`.