docs: rebuild customer-facing documentation from scratch

[moonming]

Summary

The current docs/ content is stale and no longer matches the product's real behavior. We should discard the old structure and rebuild the customer-facing documentation from scratch around the current AISIX AI Gateway and AISIX Cloud experience.

This docs set should be organized for two audiences:

• platform engineers operating and configuring the gateway
• AI agent developers integrating via OpenAI-compatible and Anthropic-style APIs

Top-level product positioning:

• AISIX AI Gateway is the primary product
• AISIX Cloud is the managed control-plane / data-plane extension

Scope

• Replace the current flat docs structure with a new information architecture
• Document only currently implemented behavior in the main docs
• Move not-yet-implemented capabilities into a dedicated docs/roadmap.md
• Treat old docs as disposable; do not preserve them for backward compatibility

New Docs Structure

• docs/index.md
• docs/roadmap.md
• docs/overview/*
• docs/quickstart/*
• docs/integration/*
• docs/configuration/*
• docs/cloud/*
• docs/operations/*
• docs/reference/*
• docs/tutorials/*

Sub-issues

• docs: create landing page, overview, feature matrix, and roadmap #245 docs: create landing page, overview, feature matrix, and roadmap
• docs: create quickstart guides for self-hosted, SDK, and AISIX Cloud #243 docs: create quickstart guides for self-hosted, SDK, and AISIX Cloud
• docs: create client integration guides for OpenAI-compatible and Anthropic-style APIs #246 docs: create client integration guides for OpenAI-compatible and Anthropic-style APIs
• docs: create gateway configuration docs for models, keys, routing, guardrails, cache, and observability #248 docs: create gateway configuration docs for models, keys, routing, guardrails, cache, and observability
• docs: create AISIX Cloud docs for environments, certificates, projection, billing, and resilience #247 docs: create AISIX Cloud docs for environments, certificates, projection, billing, and resilience
• docs: create operations docs for deployment, security, health, observability, and troubleshooting #242 docs: create operations docs for deployment, security, health, observability, and troubleshooting
• docs: create reference docs for APIs, schemas, headers, errors, and provider compatibility #244 docs: create reference docs for APIs, schemas, headers, errors, and provider compatibility
• docs: create scenario-driven tutorials for common production workflows #241 docs: create scenario-driven tutorials for common production workflows

Remaining Work Breakdown

• No page-level gaps remain in the rebuilt docs structure tracked by this parent issue.
• Remaining follow-up, if any, should be handled as doc refinements or separate behavior-specific issues rather than missing-section rebuild work.

Acceptance Criteria

• The old docs set is removed or fully superseded
• The new structure is navigable from docs/index.md
• Main docs describe only current, verified behavior
• Planned / not-yet-implemented capabilities are documented only in docs/roadmap.md
• The docs are written for customer-facing consumption, not as internal implementation notes

[coderabbitai]

🔗 Related PRs

api7/api7ee-3-control-plane#2369 - docs: add hierarchical AGENTS.md knowledge base for AI coding agents [merged]
#102 - docs: add docs about Introduction, Quick Start, Core Concepts, and Guides [merged]
api7/docs.apiseven.com#212 - docs: 新增 Anthropic 与 OpenAI 协议转换使用指南 [merged]
#8 - docs: update the architecture which is much beautiful [merged]
api7/contents#807 - seo: add 'What Is an AI Gateway' learning center article [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!

[moonming]

Rebuild complete. Closed by the merge of #249.

All 8 sub-tasks are now closed:

• docs: create landing page, overview, feature matrix, and roadmap #245 ✅ landing page, overview, feature matrix, roadmap
• docs: create quickstart guides for self-hosted, SDK, and AISIX Cloud #243 ✅ quickstarts — self-hosted + first-model-first-key-first-request shipped; remaining quickstart paths tracked in docs: add missing quickstart guides (openai-sdk, anthropic-sdk, aisix-cloud-managed-dp) #273
• docs: create client integration guides for OpenAI-compatible and Anthropic-style APIs #246 ✅ integration guides — openai-compatible-api + errors-and-retries shipped; remaining pages tracked in docs: add missing integration guides (streaming, tool-calling, anthropic-messages, embeddings, responses, audio, images, rerank, passthrough, errors-and-retries) #274
• docs: create gateway configuration docs for models, keys, routing, guardrails, cache, and observability #248 ✅ gateway configuration docs
• docs: create AISIX Cloud docs for environments, certificates, projection, billing, and resilience #247 ✅ AISIX Cloud docs
• docs: create operations docs for deployment, security, health, observability, and troubleshooting #242 ✅ operations docs
• docs: create reference docs for APIs, schemas, headers, errors, and provider compatibility #244 ✅ reference docs
• docs: create scenario-driven tutorials for common production workflows #241 ✅ scenario-driven tutorials — 4 expanded to cookbook-tier; 2 stub tutorials removed as per the rebuild's "current behavior only" rule

Also addressed: the docs API surface gap (#201).

Follow-up doc-refinement work is tracked in #273 (quickstarts) and #274 (integration pages). Any further gaps that surface will be filed as separate behavior-specific issues per the original "doc refinements rather than rebuild work" framing.