context-mode 0.5.3 → 0.5.5

package/.claude-plugin/marketplace.json

@@ -13,7 +13,7 @@
       "name": "context-mode",
       "source": "./",
       "description": "Claude Code MCP plugin that saves 98% of your context window. Sandboxed code execution in 10 languages, FTS5 knowledge base with BM25 ranking, and intent-driven search.",
-      "version": "0.5.3",
+      "version": "0.5.5",
       "author": {
         "name": "Mert Koseoğlu"
       },

package/.claude-plugin/plugin.json

@@ -1,6 +1,6 @@
 {
   "name": "context-mode",
-  "version": "0.5.3",
+  "version": "0.5.5",
   "description": "Claude Code MCP plugin that saves 98% of your context window. Sandboxed code execution in 10 languages, FTS5 knowledge base with BM25 ranking, and intent-driven search.",
   "author": {
     "name": "Mert Koseoğlu",

package/build/server.js

@@ -5,7 +5,7 @@ import { z } from "zod";
 import { PolyglotExecutor } from "./executor.js";
 import { ContentStore } from "./store.js";
 import { detectRuntimes, getRuntimeSummary, getAvailableLanguages, hasBunRuntime, } from "./runtime.js";
-const VERSION = "0.5.3";
+const VERSION = "0.5.5";
 const runtimes = detectRuntimes();
 const available = getAvailableLanguages(runtimes);
 const server = new McpServer({
@@ -389,9 +389,10 @@ server.registerTool("search", {
         "- Get configuration details ('Tailwind responsive breakpoints')\n" +
         "- Find migration steps ('App Router data fetching')\n\n" +
         "SEARCH TIPS:\n" +
-        "- Use specific technical terms, not concepts ('__proto__' not 'security')\n" +
+        "- Queries use OR semantics — results matching more terms rank higher via BM25\n" +
+        "- Use 2-4 specific technical terms per query for best results\n" +
         "- Check 'Searchable terms' from execute/execute_file results for available vocabulary\n" +
-        "- Combine multiple specific terms for better results\n\n" +
+        "- For broad topics, send multiple focused searches in parallel\n\n" +
         "Returns exact content — not summaries. Each result includes heading hierarchy and full section text.",
     inputSchema: z.object({
         query: z.string().describe("Natural language search query"),

package/build/store.js

@@ -40,7 +40,7 @@ function sanitizeQuery(query) {
         !["AND", "OR", "NOT", "NEAR"].includes(w.toUpperCase()));
     if (words.length === 0)
         return '""';
-    return words.map((w) => `"${w}"`).join(" ");
+    return words.map((w) => `"${w}"`).join(" OR ");
 }
 // ─────────────────────────────────────────────────────────
 // ContentStore

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "context-mode",
-  "version": "0.5.3",
+  "version": "0.5.5",
   "type": "module",
   "description": "Claude Code MCP plugin that saves 98% of your context window. Sandboxed code execution, FTS5 knowledge base, and intent-driven search.",
   "author": "Mert Koseoğlu",

package/skills/context-mode/SKILL.md

@@ -1,12 +1,23 @@
 ---
 name: context-mode
 description: |
-  ALWAYS use context-mode tools instead of Bash/curl/cat when output MIGHT exceed 20 lines.
-  This includes: API requests (curl, fetch, gh, aws), file reading (cat, head, tail),
-  command output (npm test, git log, docker ps, kubectl), log analysis, data processing,
-  and ANY operation where you cannot guarantee the output will be small.
-  When in doubt, use context-mode — the cost of routing through it is near zero,
-  but the cost of flooding context with 50KB+ of raw data is catastrophic.
+  Use context-mode tools (execute, execute_file, fetch_and_index, index, search) instead of
+  Bash/curl/cat when processing large outputs. The cost of routing through context-mode is near zero,
+  but flooding context with 50KB+ of raw data is catastrophic.
+
+  Use when: user wants to "fetch documentation", "look up the docs", "fetch the docs",
+  "check the API reference", "find examples in docs", "index documentation", "search docs",
+  "analyze logs", "check the logs", "read access.log", "debug the 500s",
+  "run tests", "run the test suite", "check if tests pass", "npm test", "vitest", "pytest",
+  "git log", "show recent commits", "list commits", "what changed", "diff between branches",
+  "analyze output", "summarize output", "process data", "parse JSON", "parse CSV",
+  "filter results", "extract errors", "check build output", "analyze dependencies",
+  "process API response", "large file analysis", "hit this endpoint", "call the API",
+  "check the response", "list containers", "check pods", "outdated packages",
+  "count lines", "find TODOs", "analyze codebase", "security audit",
+  or ANY operation where output might exceed 20 lines.
+  Also use when routing output from other MCP tools (Playwright snapshots, Context7 docs,
+  GitHub API responses, Sentry data) through index → search to avoid context bloat.
 ---
 
 # Context Mode: Default for All Large Output
@@ -81,6 +92,14 @@ Use context-mode for ANY of these, without being asked:
 | Shell commands with pipes | `shell` | grep, awk, jq, native tools |
 | File pattern matching | `shell` | find, wc, sort, uniq |
 
+## Search Query Strategy
+
+- BM25 uses **OR semantics** — results matching more terms rank higher automatically
+- Use 2-4 specific technical terms per query: `search("transform refine pipe")`
+- Send multiple `search()` calls **in parallel** for different aspects of a topic
+- Example: instead of one broad search, send 3 focused parallel queries:
+  - `search("transform pipe preprocess")` + `search("refine superRefine check")` + `search("coerce codec")`
+
 ## Critical Rules
 
 1. **Always console.log/print your findings.** stdout is all that enters context. No output = wasted call.