wip:RAG pipeline:general structure #6 - docs(about): write introduction, align contributing with the Python flow

Author: ByteVoyageur

docs/about/contributing.md

@@ -71,12 +71,7 @@ We rely on [Semantic Versioning](https://semver.org/) for versioning a release.
 * `MINOR` version when you add functionality in a backwards-compatible manner
 * `PATCH` version when you make backwards-compatible bug fixes.
 
-The command `npm run release:<type>`, where  `<type>` is either `patch`, `minor` or `major`, helps you to do the release. 
-
-It performs the following task for you:
-* increase the package version number in the `package.json` file
-* generate the change log
-* create a tag accordingly in the git repository and push it
+To cut a release: bump the `version` field in `pyproject.toml`, commit it, then create the matching git tag and push it (the published releases are driven by git tags).
 
 ## Contributor Code of Conduct
 

docs/about/introduction.md

@@ -7,8 +7,8 @@ sidebarDepth: 3
 <!-- TODO: 1-2 paragraphs — what knowledge is, in plain terms. -->
 
 **knowledge** is an AI developer assistant for the Kalisio ecosystem. It indexes the
-entire codebase into a vector database and exposes it to coding agents through an [MCP](https://modelcontextprotocol.io/) server, so that
-agents retrieve only the chunks relevant to a task instead of reading whole files.
+entire codebase into a vector database and serves it to coding agents through a retrieval
+API, so that agents retrieve only the chunks relevant to a task instead of reading whole files.
 
 ## The problem
 
@@ -21,12 +21,12 @@ agents retrieve only the chunks relevant to a task instead of reading whole file
 
 - **Semantic code search** — retrieval over an indexed vector database.
   See [RAG: indexing & retrieval](/architecture/rag).
-- **Git intelligence** — hotspots, co-change, bus factor.
-  See [Git intelligence](/architecture/git-intelligence).
-- **Dependency graph** — import graph and impact analysis.
-  See [Dependency graph](/architecture/dependency-graph).
-- **MCP tools** — the surface agents actually call.
-  See [MCP tools](/architecture/mcp-tools).
+- **Structure-aware ingestion** — per–file-type chunking (Markdown, JS, Vue, JSON), with
+  incremental re-indexing of only what changed. See [Ingestion pipeline](/architecture/ingestion-pipeline).
+- **Commit-history enrichment** — each chunk carries its file's recent commit history, so
+  answers reflect how a file changed. See [Ingestion pipeline](/architecture/ingestion-pipeline).
+- **Retrieval API** — `/ask` (RAG answer) and `/search` (chunks), JWT-secured.
+  See [API endpoints](/architecture/api).
 
 ## How it fits the Kalisio ecosystem
 
@@ -35,6 +35,5 @@ agents retrieve only the chunks relevant to a task instead of reading whole file
 
 ## Next steps
 
-- New here? Read [Understanding knowledge](/guides/understanding).
-- Want to run it? See [Getting started](/guides/getting-started).
 - Curious about the internals? Start with the [Architecture overview](/architecture/introduction).
+- Want the research behind the design? See [Research](/research/introduction) and [Experiments](/experiments/introduction).