docs: migrate from Sphinx to mkdocs material

Replace the Sphinx based documentation toolchain with mkdocs material to
align branding, typography and information architecture with the wider
Redis docs ecosystem.

Toolchain
- Remove docs/conf.py, docs/_extension, docs/_templates, docs/Makefile,
  docs/make.bat and the docs/_static tree
- Add mkdocs.yml configured with the material theme, Redis brand palette
  (red 500, ink 900, slate 700) and Space Grotesk / Space Mono typography
- Add scripts/mkdocs_hooks.py to suppress pre existing griffe docstring
  warnings so strict builds pass without modifying source docstrings
- Update .readthedocs.yaml to build with mkdocs
- Update Makefile docs targets (docs build, docs serve) to invoke mkdocs

Content
- Rewrite all Sphinx MyST flavored markdown to plain CommonMark plus
  pymdownx extensions
- Convert every .rst API reference page to mkdocstrings .md with the
  Google docstring style
- Render existing user_guide notebooks in place via mkdocs jupyter
- Add a Redis branded landing hero on the home page (script logo, title
  and tagline) and a redis-brand.css overrides file
- Carry over Redis favicon and architecture diagram into docs/assets

Agent affordances
- Add AGENTS.md at the repository root with a usage oriented quick
  reference for AI agents consuming redisvl
- Add docs/for-ais-only/ with REPOSITORY_MAP, BUILD_AND_TEST and
  FAILURE_MODES tailored to contributors working on the codebase
- Emit llms.txt and llms-full.txt at build time via the llmstxt plugin

Author: nkanu17

.gitignore

@@ -105,8 +105,8 @@ instance/
 # Scrapy stuff:
 .scrapy
 
-# Sphinx documentation
-docs/_build/
+# mkdocs build cache
+.cache/
 
 # PyBuilder
 .pybuilder/

.readthedocs.yaml

@@ -26,10 +26,6 @@ build:
         UV_PROJECT_ENVIRONMENT="${READTHEDOCS_VIRTUALENV_PATH}" \
           ~/.local/bin/uv sync --frozen --group docs
 
-# Build documentation in the "docs/" directory with Sphinx
-sphinx:
-  configuration: docs/conf.py
-
-formats:
-  - pdf
-  - epub
+# Build documentation with mkdocs
+mkdocs:
+  configuration: mkdocs.yml

AGENTS.md

@@ -0,0 +1,131 @@
+# AGENTS.md
+
+Quick reference for AI agents using `redisvl`. For agents working *on* the
+codebase itself, see [docs/for-ais-only/](docs/for-ais-only/index.md).
+
+## What redisvl is
+
+A Python library for using Redis as a vector database. It wraps Redis Search
+(`FT.CREATE`, `FT.SEARCH`, `FT.AGGREGATE`, vector index types) behind:
+
+- **`SearchIndex` / `AsyncSearchIndex`**: schema-driven index management.
+- **Query classes**: `VectorQuery`, `VectorRangeQuery`, `FilterQuery`,
+  `HybridQuery`, `MultiVectorQuery`, `TextQuery`, `CountQuery`, `SQLQuery`.
+- **Filter expressions**: `Tag`, `Text`, `Num`, `Geo`, `GeoRadius`.
+- **Extensions**: `SemanticCache`, `LangCacheSemanticCache`, `EmbeddingsCache`,
+  `MessageHistory` / `SemanticMessageHistory`, `SemanticRouter`.
+- **Vectorizers**: OpenAI, Azure OpenAI, Cohere, HuggingFace
+  (sentence-transformers), Mistral, Vertex AI, Bedrock, VoyageAI, custom.
+- **Rerankers**: Cohere, HuggingFace cross-encoder, VoyageAI.
+- **CLI**: `rvl index`, `rvl stats`, `rvl mcp`, `rvl version`.
+- **MCP server**: serves an existing Redis index over stdio / HTTP / SSE.
+
+## Install
+
+```bash
+pip install redisvl
+# common provider extras
+pip install redisvl[openai,cohere,sentence-transformers]
+# everything (heavy)
+pip install redisvl[all]
+```
+
+Requires Python 3.10+ and a Redis 8.x instance with the search module
+(`docker run -d -p 6379:6379 redis:8.4`).
+
+## Minimum viable use
+
+```python
+from redisvl.schema import IndexSchema
+from redisvl.index import SearchIndex
+from redisvl.query import VectorQuery
+
+schema = IndexSchema.from_dict({
+    "index": {"name": "docs", "prefix": "doc:", "storage_type": "hash"},
+    "fields": [
+        {"name": "title", "type": "text"},
+        {"name": "category", "type": "tag"},
+        {"name": "embedding", "type": "vector",
+         "attrs": {"dims": 1536, "algorithm": "hnsw",
+                   "distance_metric": "cosine", "datatype": "float32"}},
+    ],
+})
+
+index = SearchIndex(schema, redis_url="redis://localhost:6379")
+index.create(overwrite=True)
+
+index.load([
+    {"title": "intro", "category": "guide", "embedding": vector_bytes},
+])
+
+results = index.query(VectorQuery(
+    vector=query_embedding,
+    vector_field_name="embedding",
+    return_fields=["title", "category"],
+    num_results=10,
+))
+```
+
+## Public import paths (stable)
+
+Use the **subpackage**, not the module:
+
+```python
+from redisvl.index import SearchIndex, AsyncSearchIndex
+from redisvl.schema import IndexSchema
+from redisvl.query import (
+    VectorQuery, VectorRangeQuery, FilterQuery, CountQuery, TextQuery,
+    HybridQuery, MultiVectorQuery, AggregateHybridQuery, SQLQuery, Vector,
+)
+from redisvl.query.filter import Tag, Text, Num, Geo, GeoRadius
+from redisvl.extensions.cache.llm import SemanticCache, LangCacheSemanticCache
+from redisvl.extensions.message_history import (
+    MessageHistory, SemanticMessageHistory,
+)
+from redisvl.extensions.router import SemanticRouter, Route, RoutingConfig
+from redisvl.utils.vectorize import (
+    HFTextVectorizer, OpenAITextVectorizer, AzureOpenAITextVectorizer,
+    CohereTextVectorizer, MistralAITextVectorizer, VoyageAIVectorizer,
+    VertexAIVectorizer, BedrockVectorizer, CustomVectorizer,
+)
+from redisvl.utils.rerank import (
+    CohereReranker, HFCrossEncoderReranker, VoyageAIReranker,
+)
+```
+
+## What docs to read
+
+- Concepts → [docs/concepts/](docs/concepts/index.md): how indexes, schemas,
+  queries, and extensions fit together.
+- User Guide → [docs/user_guide/](docs/user_guide/index.md): notebooks for
+  every common task.
+- API Reference → [docs/api/](docs/api/index.md): the generated reference.
+- Examples → [docs/examples/](docs/examples/index.md): links to
+  redis-ai-resources for end-to-end recipes.
+- For AI Agents (codebase contributors) →
+  [docs/for-ais-only/](docs/for-ais-only/index.md).
+
+## Machine-readable indexes
+
+When the docs are built, they emit:
+
+- [`llms.txt`](https://docs.redisvl.com/llms.txt) — flat index of every doc.
+- [`llms-full.txt`](https://docs.redisvl.com/llms-full.txt) — concatenated
+  full content for one-shot loading.
+
+## Things to know before suggesting code
+
+- **Always combine schema + algorithm changes.** Bundling datatype and
+  algorithm changes into a single index patch produces one drop/rebuild cycle
+  instead of two.
+- **`MessageHistory` / `SemanticMessageHistory`** replace the deprecated
+  `SessionManager` / `SemanticSessionManager`. The old names still import but
+  emit a `DeprecationWarning` and will be removed.
+- **SVS-VAMANA** requires Redis ≥ 8.2.0 with Redis Search ≥ 2.8.10 and only
+  supports `float16` / `float32` datatypes.
+- **`SQLQuery`** requires the `redisvl[sql-redis]` extra and translates SQL
+  `SELECT` into `FT.SEARCH` / `FT.AGGREGATE` via the
+  [sql-redis](https://github.com/redis-developer/sql-redis) project.
+- **`HybridQuery` vs `AggregateHybridQuery`** weight scores differently:
+  `HybridQuery.linear_alpha` weights *text*, `AggregateHybridQuery.alpha`
+  weights *vector*. Recheck `alpha` when switching.

Makefile

@@ -73,12 +73,11 @@ check: lint test ## Run all checks (lint + test)
 
 docs-build: ## Build documentation
 	@echo "📚 Building documentation"
-	uv run make -C docs html
+	uv run mkdocs build --strict
 
 docs-serve: ## Serve documentation locally
-	@echo "🌐 Serving documentation at http://localhost:8000"
-	@echo "📁 Make sure docs are built first with 'make docs-build'"
-	uv run python -m http.server --directory docs/_build/html
+	@echo "🌐 Serving documentation at http://127.0.0.1:8000"
+	uv run mkdocs serve --dev-addr 127.0.0.1:8000
 
 build: ## Build wheel and source distribution
 	@echo "🏗️ Building distribution packages"