codesearch-mcp 0.0.17-alpha → 0.0.17-alpha.1

package/README.md

@@ -1,8 +1,8 @@
 # codesearch-mcp
 
-Semantic codebase search for Claude. Indexes your local projects into a PostgreSQL vector database and exposes search via the Model Context Protocol (MCP).
+Semantic codebase search for AI agents. Indexes your local projects into a PostgreSQL vector database and exposes search via the Model Context Protocol (MCP).
 
-Ask Claude things like *"where is the payment retry logic?"* or *"show me how auth tokens are validated"* — across all your projects at once.
+Ask your AI agent things like *"where is the payment retry logic?"* or *"show me how auth tokens are validated"* — across all your projects at once.
 
 > **Alpha** — works well locally, API may change before 1.0.
 
@@ -107,25 +107,6 @@ Pass `include_graph_context: true` to `search_codebase` to see direct imports an
 
 Useful for immediately understanding a file's role in the dependency graph.
 
-## Cross-encoder reranking
-
-By default, results are ordered by cosine distance (vector similarity). Enable cross-encoder reranking to reorder candidates by semantic relevance — better for queries where the exact wording differs from how the code is written.
-
-Requires Ollama ≥ 0.6 and a reranker model:
-
-```bash
-ollama pull qllama/bge-reranker-v2-m3
-```
-
-```bash
-rag-config set reranker.enabled true
-rag-config set reranker.model qllama/bge-reranker-v2-m3
-# rag-config set reranker.url http://localhost:11434/api/rerank  # default
-# rag-config set reranker.overfetch 3                            # fetch limit×3 candidates before reranking
-```
-
-When enabled, `search_codebase` fetches `limit × reranker.overfetch` vector candidates, then the cross-encoder reranks them and returns the top `limit`. Falls back gracefully to vector order if the reranker is unreachable.
-
 ## Excluding files
 
 Create `.ragignore` in a project root — same syntax as `.gitignore`:
@@ -154,7 +135,14 @@ rag-config set logging.dir false
 
 ## Configuration
 
-All set by `rag-setup`. Override individual keys with `rag-config set <key> <value>`. Stored in `~/.config/codesearch-mcp/config.json`.
+All set by `rag-setup`. Stored in `~/.config/codesearch-mcp/config.json`.
+
+```bash
+rag-config show                      # print current config (password redacted)
+rag-config get <key>                 # print a single value
+rag-config set <key> <value>         # update a value
+rag-config reset <key>               # remove key (reverts to default)
+```
 
 | Key | Default | Description |
 |-----|---------|-------------|
@@ -170,10 +158,6 @@ All set by `rag-setup`. Override individual keys with `rag-config set <key> <val
 | `search.autoTranslate` | `false` | Translate non-English queries to English before embedding |
 | `search.translateModel` | `llama3.2` | Ollama model for translation (requires `ollama pull llama3.2`) |
 | `indexing.astBatchSize` | `3000` | Max files per ts-morph batch — lower if indexing large projects causes OOM |
-| `reranker.enabled` | `false` | Enable cross-encoder reranking (requires Ollama ≥ 0.6 and `reranker.model`) |
-| `reranker.model` | — | Reranker model name (e.g. `qllama/bge-reranker-v2-m3`) |
-| `reranker.url` | `${ollama.host}/api/rerank` | Override rerank endpoint URL |
-| `reranker.overfetch` | `3` | Candidate multiplier before reranking (`limit × N`, max 20) |
 | `logging.level` | `error` | Log level written to file: `error` \| `warn` \| `info` \| `debug` |
 | `logging.dir` | `~/.config/codesearch-mcp/logs` | Base log directory, or `"false"` to disable |
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "codesearch-mcp",
-  "version": "0.0.17-alpha",
+  "version": "0.0.17-alpha.1",
   "description": "Local semantic codebase search via MCP — indexes projects into pgvector, exposes search to Claude",
   "license": "MIT",
   "main": "dist/app.js",