xindex 1.0.4 → 1.0.5

package/README.md

@@ -39,6 +39,15 @@ Drop this into `.mcp.json` at your project root:
 
 Open the project in Claude Code — it picks up the xindex MCP server and can call `xindex_search`, `xindex_index`, and `xindex_reset` directly. Fewer hallucinations, fewer round-trips.
 
+## Features
+
+- **Local** — everything runs on your machine; embeddings cached on disk
+- **Semantic search** — natural-language queries, not substring match
+- **MCP server** — plugs into Claude Code via `.mcp.json`
+- **Watch mode** — keeps the index warm while you code
+- **Gitignore-aware** — respects `.gitignore` + custom ignore rules
+- **Zero config** — works with defaults; `.xindex.json` is optional
+
 ## Claude Code skills (`@xi`)
 
 Two optional [Claude Code skills](https://docs.claude.com/en/docs/claude-code/skills) wrap the MCP tools so you don't have to think about them:
@@ -73,13 +82,12 @@ argument-hint: "[question]"
 Surface-level codebase discovery via xindex. Tool: `xindex_search` (natural-language, meaning-based).
 
 **Steps:**
-1. Draft 5–10 focused queries from $ARGUMENTS (entry points, routing, config, integrations, tests, related patterns).
-2. Run `xindex_search` for each.
-3. If results are empty/sparse/stale → scoped-index the most relevant content-heavy root folders (one path per `xindex_index` call, e.g. `src`, `apps`, `features`, `componets`), then re-search. Prefer scoped over full-repo.
-4. Refine with 2–3 narrower follow-ups.
-5. Return file paths + brief keywords showing why each matched.
+1. Draft 3–7 queries from $ARGUMENTS (entry points, routing, config, tests, patterns); run `xindex_search` in parallel.
+2. If empty/sparse/stale → scoped `xindex_index` on relevant root folders (one path per call, e.g. `src`, `skills`, `agents`), then re-search. Prefer scoped over full-repo.
+3. Run 3–7 narrower follow-ups in parallel based on round-1 hits.
+4. Return file paths + brief keywords showing why each matched.
 
-Output = file links + keywords, not analysis. For reset or full re-index, delegate to `/xindex` (owns safety rules).
+Output = file links + keywords, not analysis. **Escalate to `/ask-cursor` by default** (cheap codebase reasoning); only go to `/ask-claude` for multi-file/pattern analysis or `/ask-claude-opus` for trade-offs. For reset or full re-index, delegate to `/xindex` (owns safety rules).
 ````
 
 `xindex/SKILL.md`:
@@ -90,7 +98,7 @@ name: xindex
 description: Manages xindex semantic search — index, search, reset via MCP tools. For research questions, use /ask-xi.
 argument-hint: "[search query | index | reset]"
 ---
-Full xindex tool management. For research, use `/ask-xi`. Install: `npm i -g xindex`.
+Full xindex tool management. For research, use `/ask-xi`. Install: `npm i -g xindex` ([npm](https://www.npmjs.com/package/xindex)).
 
 **Tools:**
 - `xindex_search` — find files by meaning (synonyms, semantics). Try before grepping blindly.
@@ -106,15 +114,6 @@ $ARGUMENTS
 
 Both skills assume the `xindex` MCP server is registered (see the section above). Restart Claude Code after adding skills.
 
-## Features
-
-- **Local** — everything runs on your machine; embeddings cached on disk
-- **Semantic search** — natural-language queries, not substring match
-- **MCP server** — plugs into Claude Code via `.mcp.json`
-- **Watch mode** — keeps the index warm while you code
-- **Gitignore-aware** — respects `.gitignore` + custom ignore rules
-- **Zero config** — works with defaults; `.xindex.json` is optional
-
 ---
 
 ## CLI reference
@@ -129,7 +128,7 @@ xindex-index apps features
 ```
 
 ### `xindex-search <query...>`
-Search the index. All args are joined into one query. Default limit: 10.
+Search the index. All args are joined into one query. Default limit: 7.
 ```bash
 xindex-search "database migration logic"
 xindex-search file watcher debounce
@@ -157,19 +156,22 @@ xindex-mcp --watch-dir=./src # watch a specific dir
 
 ## MCP tools
 
-| Tool | What it does | Input |
-|------|--------------|-------|
-| `xindex_search` | Semantic search | `query: string`, `limit?: number` (default 5, max 100) |
-| `xindex_index`  | Index paths | `inputs: string[]` (at least one) |
-| `xindex_reset`  | Wipe index (destructive) | — |
+- **`xindex_search`** — semantic search. `query: string`, `limit?: number` (default 7, max 50)
+- **`xindex_index`** — index paths. `inputs: string[]` (at least one)
+- **`xindex_reset`** — wipe index (destructive). No input
 
-Note: CLI `xindex-search` defaults to 10 results; MCP `xindex_search` defaults to 5.
+Note: both CLI `xindex-search` and MCP `xindex_search` default to 7 results; MCP caps at 50.
 
 ## Configuration
 
 ### `.xindex.json` (optional)
 
-Place at your project root. Both fields are optional arrays; unknown keys are ignored.
+Project-root file. All fields optional; unknown keys ignored; missing/empty → defaults.
+
+- **`ignoreKeywords`** — `string[]`, default `[]`. Tokens stripped before embedding — add project slang/boilerplate polluting results. Entries ≤1 char warn.
+- **`ignoreFiles`** — `string[]`, default `[]`. Extra globs excluded during walk/watch, on top of `.gitignore` — add vendored/generated folders.
+- **`maxLines`** — `number`, default `30`. Lines per chunk — tune if chunks feel over/under-sized.
+- **`maxFileBytes`** — `number`, default `5000000`. Skip files over this (5 MB) — lower for faster indexing on huge generated files.
 
 ```json
 {
@@ -178,8 +180,7 @@ Place at your project root. Both fields are optional arrays; unknown keys are ig
 }
 ```
 
-- **`ignoreKeywords`** — tokens stripped before embedding (noise words, language keywords, project slang). Keeps search focused on meaningful terms.
-- **`ignoreFiles`** — extra glob patterns excluded during walk/watch, on top of `.gitignore`.
+Override only what you need; re-run `xindex-index .` (or let the watcher pick it up). Invalid JSON throws; wrong-typed fields fall back silently.
 
 ### `.xindex/` folder
 

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "xindex",
-  "version": "1.0.4",
+  "version": "1.0.5",
   "description": "Local semantic code search — index codebase, search by meaning or keywords",
   "type": "module",
   "main": "xindex.ts",