Tighten docs voice and update docs URL

Author: auscaster

CHANGELOG.md

@@ -1,5 +1,10 @@
 # Changelog
 
+## 0.1.3 - 2026-04-24
+
+- pointed package documentation at a dedicated Sourcey docs guide
+- tightened package copy and removed the download badge
+
 ## 0.1.2 - 2026-04-24
 
 - improved the package README and added runnable example scripts

README.md

@@ -1,37 +1,32 @@
 # langchain-sourcey
 
 [![PyPI - Version](https://img.shields.io/pypi/v/langchain-sourcey?label=%20)](https://pypi.org/project/langchain-sourcey/)
-[![PyPI - Downloads](https://img.shields.io/pepy/dt/langchain-sourcey)](https://pypistats.org/packages/langchain-sourcey)
 [![PyPI - License](https://img.shields.io/pypi/l/langchain-sourcey)](https://pypi.org/project/langchain-sourcey/)
 [![CI](https://github.com/sourcey/langchain-sourcey/actions/workflows/ci.yml/badge.svg)](https://github.com/sourcey/langchain-sourcey/actions/workflows/ci.yml)
 
-`langchain-sourcey` is the native LangChain retriever for Sourcey-generated
-documentation sites.
+> Your docs retriever should not depend on somebody else's SaaS either.
 
-It turns a published Sourcey docs root into a LangChain knowledge source
-without a private indexing service or ingestion pipeline. The retriever works
-directly against Sourcey's public artefacts:
+`langchain-sourcey` reads a published Sourcey docs site directly.
+
+Sourcey already ships the files a retriever needs. This package uses them:
 
 - `search-index.json` for candidate discovery
 - `llms-full.txt` for full-page hydration
 - canonical page URLs for citations
 
-## Why this integration is a good LangChain fit
-
-- No credentials required for public docs sites
-- Retrieval works against static hosting, subpath deployments, and GitHub Pages
-- Returned `Document` objects carry canonical `metadata["source"]` URLs
-- `llms-full.txt` gives cleaner full-page content than scraping rendered HTML
-- If `llms-full.txt` is missing, the retriever falls back to page HTML
+If `llms-full.txt` is missing, it falls back to the matched page HTML.
 
 ## Install
 
 ```bash
 pip install -U langchain-sourcey
 ```
 
-Point `site_url` at the root of a published Sourcey build, for example
-`https://sourcey.com/docs` or `https://sourcey.com/cheesestore`.
+Point `site_url` at the root of a published Sourcey build:
+
+- `https://sourcey.com/docs`
+- `https://sourcey.com/cheesestore`
+- `https://cheesestore.github.io`
 
 ## Quickstart
 
@@ -54,6 +49,8 @@ for doc in docs:
 
 For a runnable script, see [examples/live_quickstart.py](examples/live_quickstart.py).
 
+Sourcey guide: `https://sourcey.com/docs/guides/guide-langchain-retriever.html`
+
 ## Use In A LangChain Chain
 
 Install a chat model integration of your choice. This example uses OpenAI:
@@ -93,17 +90,16 @@ print(answer)
 
 For a fuller example, see [examples/rag_chain.py](examples/rag_chain.py).
 
-## Sourcey Site Contract
+## Sourcey Contract
 
-For best results, the published Sourcey site should:
+This package assumes the published Sourcey site exposes:
 
 - publish `search-index.json`
 - publish `llms-full.txt`
 - set `siteUrl` in `sourcey.config.ts` so citations are canonical
 
 `search-index.json` is required. `llms-full.txt` is strongly recommended because
-it lets the retriever return full page content instead of HTML-derived fallback
-text.
+it gives the retriever full page content instead of HTML-derived fallback text.
 
 ## Returned Metadata
 
@@ -141,6 +137,4 @@ This repo includes draft docs ready to turn into a LangChain docs PR:
 
 ## Scope
 
-This package intentionally ships `SourceyRetriever` only. A document loader is
-deferred until the retriever proves enough demand to justify the maintenance
-surface.
+This package ships `SourceyRetriever` only. No loader yet.

docs/langchain/provider-sourcey.mdx

@@ -4,21 +4,20 @@ description: "Integrate with Sourcey using LangChain Python."
 ---
 
 > [Sourcey](https://sourcey.com) turns documentation, API references, and MCP
-> servers into static sites built for both humans and LLMs.
-> `langchain-sourcey` lets LangChain retrieve directly from those published
-> Sourcey artefacts.
+> servers into static sites.
+> `langchain-sourcey` reads those published artefacts directly.
 
 ## Installation and setup
 
-`Sourcey` integration exists in its own partner package. Install it with:
+Install `langchain-sourcey`:
 
 ```bash
 pip install -qU langchain-sourcey
 ```
 
 No API key is required. Point the retriever at a public Sourcey docs root that
-publishes `search-index.json`. `llms-full.txt` is recommended for full-page
-hydration and stable citations.
+publishes `search-index.json`. If `llms-full.txt` is there, the retriever uses
+it for full page content.
 
 ## Retriever
 

docs/langchain/retriever-sourcey.mdx

@@ -4,14 +4,14 @@ description: "Integrate with the Sourcey retriever using LangChain Python."
 ---
 
 > [Sourcey](https://sourcey.com) publishes static documentation sites that
-> already contain the retrieval artefacts LangChain needs.
+> already contain the retrieval artefacts this package needs.
 > `SourceyRetriever` searches `search-index.json`, hydrates content from
 > `llms-full.txt`, and returns `Document` objects with canonical page URLs in
 > `metadata["source"]`.
 
 ## Setup
 
-Install the partner package:
+Install the package:
 
 ```bash
 pip install -qU langchain-sourcey

pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "langchain-sourcey"
-version = "0.1.2"
+version = "0.1.3"
 description = "Native LangChain retriever for Sourcey-generated documentation sites."
 readme = "README.md"
 license = { file = "LICENSE" }
@@ -35,7 +35,7 @@ dev = [
 
 [project.urls]
 Homepage = "https://sourcey.com"
-Documentation = "https://github.com/sourcey/langchain-sourcey#readme"
+Documentation = "https://sourcey.com/docs/guides/guide-langchain-retriever.html"
 Repository = "https://github.com/sourcey/langchain-sourcey"
 Issues = "https://github.com/sourcey/langchain-sourcey/issues"
 Changelog = "https://github.com/sourcey/langchain-sourcey/blob/main/CHANGELOG.md"

src/langchain_sourcey/retrievers.py

@@ -92,7 +92,7 @@ class SourceyRetriever(BaseRetriever):
     top_k: int = Field(default=6, gt=0)
     timeout: float = Field(default=10.0, gt=0)
     use_llms_full: bool = True
-    user_agent: str = "langchain-sourcey/0.1.2"
+    user_agent: str = "langchain-sourcey/0.1.3"
     headers: dict[str, str] = Field(default_factory=dict)
     transport: httpx.BaseTransport | None = Field(default=None, exclude=True, repr=False)