# Caesar

> Agent-first search API with provenance: search, read, and feedback across HTTP, SDKs, CLI, MCP, and Skills.

## Docs

- [Agents guide](https://docs.trycaesar.com/agents/agents-guide.md): Which Caesar surface to use from which agent harness, and the conventions that hold across all of them.
- [Install for agents](https://docs.trycaesar.com/agents/install.md): Copy-paste Caesar install blocks for every agent environment, each with a verification step.
- [Get a document](https://docs.trycaesar.com/api-reference/documents/get-a-document.md): Inspect one canonical document and retrieve selected content.
- [Record feedback](https://docs.trycaesar.com/api-reference/feedback/record-feedback.md): Persist an agent or evaluation feedback event.
- [API reference](https://docs.trycaesar.com/api-reference/index.md): Base URL, authentication, conventions, and the three endpoints of the Caesar search API.
- [Search](https://docs.trycaesar.com/api-reference/search/search.md): Run ranked retrieval over canonical documents and passages.
- [Authentication](https://docs.trycaesar.com/authentication.md): Bearer API keys, the anonymous tier, scopes, and key-handling conventions for the Caesar API.
- [Scripting and CI](https://docs.trycaesar.com/clients/cli/automation.md): Use the caesar-search CLI in scripts, schedulers, and CI pipelines: JSON-only parsing, exit-code control flow, configuration precedence, and the raw API escape hatch.
- [Install the CLI](https://docs.trycaesar.com/clients/cli/install.md): Install the caesar-search CLI with Homebrew, npm, or the standalone installer, then authenticate and keep it up to date.
- [CLI usage](https://docs.trycaesar.com/clients/cli/usage.md): Command reference for the caesar-search CLI: search, read, feedback, JSON output, and the exit-code contract.
- [Agent Development Kit](https://docs.trycaesar.com/clients/integrations/adk.md): Use Caesar as function tools or MCP tools in Agent Development Kit projects.
- [CrewAI](https://docs.trycaesar.com/clients/integrations/crewai.md): Create CrewAI tools that expose Caesar search and document reading to multi-agent workflows.
- [LangChain](https://docs.trycaesar.com/clients/integrations/langchain.md): Wrap Caesar's Python or TypeScript client as LangChain tools for search and document reading.
- [LlamaIndex](https://docs.trycaesar.com/clients/integrations/llamaindex.md): Use Caesar as LlamaIndex function tools for search-first, read-second retrieval workflows.
- [Tool calling](https://docs.trycaesar.com/clients/integrations/tool-calling.md): Expose Caesar search and read as generic function tools in any model runtime that supports tool calls.
- [Remote MCP server](https://docs.trycaesar.com/clients/mcp/remote.md): Connect any MCP client to Caesar's hosted streamable-HTTP server for web search and document reading.
- [Vercel AI SDK tools](https://docs.trycaesar.com/clients/sdks/ai-sdk.md): Give any Vercel AI SDK agent Caesar search and read tools with one import.
- [Python SDK](https://docs.trycaesar.com/clients/sdks/python.md): Typed Python client for the Caesar API with sync and async support, automatic retries, and pydantic v2 models.
- [TypeScript SDK](https://docs.trycaesar.com/clients/sdks/typescript.md): Install and use the caesar-search package for Node, Bun, Deno, and edge runtimes.
- [Agent Skills](https://docs.trycaesar.com/clients/skills.md): Install Caesar's SKILL.md files so your agent learns the search, read, and feedback loop — served directly by the API, with curl and npx skills options.
- [Documents](https://docs.trycaesar.com/concepts/documents.md): How POST /v1/document works: doc_id and canonical_url lookups, content selection, continuation reads, capture history, and passages.
- [Errors and warnings](https://docs.trycaesar.com/concepts/errors.md): The error envelope, every error and warning code the API emits, retry guidance, and the CLI exit-code contract.
- [How search works](https://docs.trycaesar.com/concepts/how-search-works.md): How Caesar turns the public web into canonical documents, passages, rankings, and provenance handles for agents.
- [Provenance](https://docs.trycaesar.com/concepts/provenance.md): Caesar attaches stable identity handles to every result so agents can cite exactly what they read, re-read it, and detect when it changes.
- [Response shaping](https://docs.trycaesar.com/concepts/response-shaping.md): Choose how much of each search result the API serializes — verbosity presets and character budgets, decided per call by the caller.
- [Search](https://docs.trycaesar.com/concepts/search.md): How POST /v1/search works: retrieval modes, query variants, result limits, filters, freshness and source policies, and the ranking contract.
- [Sessions](https://docs.trycaesar.com/concepts/sessions.md): Group a multi-step task with a session_id so feedback and ranking signals connect across calls.
- [What is Caesar](https://docs.trycaesar.com/index.md): Caesar is an agent-first search API where every result carries stable, citable handles you can read in full and feed back into ranking.
- [Pricing](https://docs.trycaesar.com/pricing.md): Caesar search is free to use, with optional partner API keys for higher throughput and attribution.
- [Connect Caesar to your agent](https://docs.trycaesar.com/quickstart.md): Install Caesar search into an agent, verify search and read, then see the raw API loop with no API key.
- [Rate limits](https://docs.trycaesar.com/rate-limits.md): Per-second rate limits by tier, the X-RateLimit response headers, and how to back off on 429s.

## OpenAPI Specs

- [openapi](https://docs.trycaesar.com/api-reference/openapi.json)