ctx7 0.4.0 → 0.4.1

package/dist/index.js

@@ -453,11 +453,9 @@ async function getLibraryContext(libraryId, query, options, accessToken) {
   if (options?.type) {
     params.set("type", options.type);
   }
-  if (options?.researchMode) {
-    params.set("researchMode", "true");
-  }
+  const headers = getAuthHeaders(accessToken);
   const response = await fetch(`${baseUrl}/api/v2/context?${params}`, {
-    headers: getAuthHeaders(accessToken)
+    headers
   });
   if (!response.ok) {
     const errorData = await response.json().catch(() => ({}));
@@ -2783,8 +2781,7 @@ Do not use for: refactoring, writing scripts from scratch, debugging business lo
 1. \`resolve-library-id\` with the library name and the user's question. Use the official library name with proper punctuation (e.g., "Next.js" not "nextjs", "Customer.io" not "customerio", "Three.js" not "threejs")
 2. Pick the best match by: exact name match, description relevance, code snippet count, source reputation (High/Medium preferred), and benchmark score (higher is better). Use version-specific IDs when the user mentions a version
 3. \`query-docs\` with the selected library ID and the user's full question (not single words)
-4. If you weren't satisfied with the answer, call \`query-docs\` again for the same library with \`researchMode: true\`. This retries with sandboxed agents that git-pull the actual source repos plus a live web search, then synthesizes a fresh answer. More costly than the default
-5. Answer using the fetched docs
+4. Answer using the fetched docs
 `;
 var FALLBACK_CLI = `Use the \`ctx7\` CLI to fetch current documentation whenever the user asks about a library, framework, SDK, API, CLI tool, or cloud service -- even well-known ones like React, Next.js, Prisma, Express, Tailwind, Django, or Spring Boot. This includes API syntax, configuration, version migration, library-specific debugging, setup instructions, and CLI tool usage. Use even when you think you know the answer -- your training data may not reflect recent changes. Prefer this over web search for library docs.
 
@@ -2795,8 +2792,7 @@ Do not use for: refactoring, writing scripts from scratch, debugging business lo
 1. Resolve library: \`npx ctx7@latest library <name> "<user's question>"\`
 2. Pick the best match by: exact name match, description relevance, code snippet count, source reputation (High/Medium preferred), and benchmark score (higher is better). If results don't look right, try the full name with punctuation (e.g., "next.js" not "nextjs")
 3. Fetch docs: \`npx ctx7@latest docs <libraryId> "<user's question>"\`
-4. If you weren't satisfied with the answer, re-run the same command with \`--research\`. This retries with sandboxed agents that git-pull the actual source repos plus a live web search, then synthesizes a fresh answer. More costly than the default
-5. Answer using the fetched documentation
+4. Answer using the fetched documentation
 
 You MUST call \`library\` first to get a valid ID (format: \`/org/project\`) unless the user provides one directly. Use the user's full question as the query -- specific and detailed queries return better results than vague single words. Do not run more than 3 commands per question. Do not include sensitive information (API keys, passwords, credentials) in queries.
 
@@ -3806,19 +3802,12 @@ async function queryCommand(libraryId, query, options) {
     process.exitCode = 1;
     return;
   }
-  const spinner = isTTY ? ora6(
-    options.research ? `Researching "${libraryId}"...` : `Fetching docs for "${libraryId}"...`
-  ).start() : null;
   const accessToken = getAccessToken();
+  const spinner = isTTY ? ora6(`Fetching docs for "${libraryId}"...`).start() : null;
   const outputType = options.json ? "json" : "txt";
   let result;
   try {
-    result = await getLibraryContext(
-      libraryId,
-      query,
-      { type: outputType, researchMode: options.research },
-      accessToken
-    );
+    result = await getLibraryContext(libraryId, query, { type: outputType }, accessToken);
   } catch (err) {
     spinner?.fail(`Error: ${err instanceof Error ? err.message : String(err)}`);
     if (!spinner) log.error(err instanceof Error ? err.message : String(err));
@@ -3882,14 +3871,9 @@ function registerDocsCommands(program2) {
   program2.command("library").argument("<name>", "Library name to search for").argument("[query]", "Question or task for relevance ranking").option("--json", "Output as JSON").description("Resolve a library name to a Context7 library ID").action(async (name, query, options) => {
     await resolveCommand(name, query, options);
   });
-  program2.command("docs").argument("<libraryId>", "Context7 library ID (e.g., /facebook/react)").argument("<query>", "Question or task to get docs for").option("--json", "Output as JSON").option(
-    "--research",
-    "Retry the query with deep research: spins up sandboxed agents that read the actual source repos (git pull + inspect) and runs a live web search, then synthesizes a fresh answer. Use on retry if you weren't satisfied with the first answer. More costly than the default."
-  ).description("Query documentation for a library").action(
-    async (libraryId, query, options) => {
-      await queryCommand(libraryId, query, options);
-    }
-  );
+  program2.command("docs").argument("<libraryId>", "Context7 library ID (e.g., /facebook/react)").argument("<query>", "Question or task to get docs for").option("--json", "Output as JSON").description("Query documentation for a library").action(async (libraryId, query, options) => {
+    await queryCommand(libraryId, query, options);
+  });
 }
 
 // src/commands/upgrade.ts