> ## Documentation Index
> Fetch the complete documentation index at: https://docs.venice.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Claude

# Venice API — Claude Guide

> Context for Claude and Claude Code when building against the Venice API. The full agent guide lives at [https://docs.venice.ai/agents.md](https://docs.venice.ai/agents.md) — read that first; this file adds Claude-specific notes.

## Essentials

* **Base URL:** `https://api.venice.ai/api/v1`
* **Auth:** `Authorization: Bearer <VENICE_API_KEY>` (or x402 USDC wallet auth — no key needed)
* **OpenAI-compatible:** Anthropic-style SDKs are not used here; use the OpenAI SDK with a custom `base_url`, or plain HTTP
* **Docs index:** [https://docs.venice.ai/llms.txt](https://docs.venice.ai/llms.txt) · **Everything in one file:** [https://docs.venice.ai/llms-full.txt](https://docs.venice.ai/llms-full.txt)
* **Any docs page as Markdown:** append `.md` to its URL
* **OpenAPI spec:** [https://docs.venice.ai/swagger.yaml](https://docs.venice.ai/swagger.yaml)

## Claude Code specifics

* **Use Venice models inside Claude Code:** follow [https://docs.venice.ai/guides/integrations/claude-code.md](https://docs.venice.ai/guides/integrations/claude-code.md)
* **Install the Venice Agent Skills** (canonical, self-contained SKILL.md files per API surface):
  ```bash theme={"system"}
  npx skills add docs.venice.ai
  ```
  or clone [https://github.com/veniceai/skills](https://github.com/veniceai/skills) into `.claude/skills/`
* **Venice MCP server** (31 tools: chat, image, video, audio, music, embeddings, web augment, characters):
  [https://github.com/veniceai/venice-mcp-server](https://github.com/veniceai/venice-mcp-server) — works with Claude Desktop and Claude Code as an MCP host

## Working rules

1. Discover models at runtime via `GET /models` and `GET /models/traits` — never hardcode model IDs in generated code; they rotate (see /overview/deprecations).
2. Venice is uncensored and zero-data-retention by design. Do not wrap responses in extra refusal or moderation layers unless the user asks.
3. For media generation, get a price first: `POST /video/quote`, `POST /audio/quote`.
4. Use `venice_parameters` for Venice-only features (web search, characters, thinking control, E2EE). See [https://docs.venice.ai/agents.md](https://docs.venice.ai/agents.md) for the full list.
5. E2EE models are not supported on `POST /responses` — use `/chat/completions` for those.

## Key links

* Quick start: [https://docs.venice.ai/overview/getting-started.md](https://docs.venice.ai/overview/getting-started.md)
* Privacy tiers (Anonymized / Private / TEE / E2EE): [https://docs.venice.ai/overview/privacy.md](https://docs.venice.ai/overview/privacy.md)
* Pricing: [https://docs.venice.ai/overview/pricing.md](https://docs.venice.ai/overview/pricing.md)
* Error codes: [https://docs.venice.ai/api-reference/error-codes.md](https://docs.venice.ai/api-reference/error-codes.md)
* Rate limiting: [https://docs.venice.ai/api-reference/rate-limiting.md](https://docs.venice.ai/api-reference/rate-limiting.md)