docs: add CHANGELOG entry for v1.0.63

feat: add pagination support for CI job logs to prevent context window flooding

CHANGELOG.md

@@ -1,3 +1,29 @@
+## [1.0.63] - 2025-06-12
+
+### Added
+
+- 📊 **CI Job Log Pagination**: Added pagination support for CI job logs to prevent context window flooding
+  - `get_pipeline_job_output` now supports optional `limit` and `offset` parameters
+  - Default limit is 1000 lines when pagination is used
+  - Returns lines from the end of the log, with configurable offset
+  - Includes truncation metadata showing what was skipped
+  - Maintains backward compatibility (no parameters = full log)
+  - See: [PR #97](https://github.com/zereight/gitlab-mcp/pull/97)
+
+---
+
+## [1.0.62] - 2025-06-10
+
+### Fixed
+
+- 🔐 **Private Token Authentication Fix**: Fixed Private-Token header authentication for GitLab API
+  - Removed incorrect `Bearer ` prefix from Private-Token header in legacy authentication mode
+  - Fixed authentication issues when using older GitLab instances with private tokens
+  - Ensures proper API authentication for both new and legacy GitLab configurations
+  - See: [PR #91](https://github.com/zereight/gitlab-mcp/pull/91), [Issue #88](https://github.com/zereight/gitlab-mcp/issues/88)
+
+---
+
 ## [1.0.60] - 2025-06-07
 
 ### Added

RELEASE_NOTES.md

@@ -1,18 +1,11 @@
-## [1.0.60] - 2025-06-07
-
-### Added
-
-- 📄 **Merge Request Enhancement**: Added support for `remove_source_branch` and `squash` options for merge requests
-  - Enhanced merge request functionality with additional configuration options
-  - Allows automatic source branch removal after merge
-  - Supports squash commits for cleaner Git history
-  - See: [PR #86](https://github.com/zereight/gitlab-mcp/pull/86)
+## [1.0.62] - 2025-06-10
 
 ### Fixed
 
-- 🔧 **Issue Assignment Fix**: Fixed list issues assignee username handling
-  - Corrected assignee username field in issue listing functionality
-  - Improved user assignment data processing for GitLab issues
-  - See: [PR #87](https://github.com/zereight/gitlab-mcp/pull/87), [Issue #74](https://github.com/zereight/gitlab-mcp/issues/74)
+- 🔐 **Private Token Authentication Fix**: Fixed Private-Token header authentication for GitLab API
+  - Removed incorrect `Bearer ` prefix from Private-Token header in legacy authentication mode
+  - Fixed authentication issues when using older GitLab instances with private tokens
+  - Ensures proper API authentication for both new and legacy GitLab configurations
+  - See: [PR #91](https://github.com/zereight/gitlab-mcp/pull/91), [Issue #88](https://github.com/zereight/gitlab-mcp/issues/88)
 
 ---

index.ts

@@ -523,7 +523,7 @@ const allTools = [
   },
   {
     name: "get_pipeline_job_output",
-    description: "Get the output/trace of a GitLab pipeline job number",
+    description: "Get the output/trace of a GitLab pipeline job with optional pagination to limit context window usage",
     inputSchema: zodToJsonSchema(GetPipelineJobOutputSchema),
   },
   {
@@ -2611,9 +2611,11 @@ async function getPipelineJob(projectId: string, jobId: number): Promise<GitLabP
  *
  * @param {string} projectId - The ID or URL-encoded path of the project
  * @param {number} jobId - The ID of the job
+ * @param {number} limit - Maximum number of lines to return from the end (default: 1000)
+ * @param {number} offset - Number of lines to skip from the end (default: 0)
  * @returns {Promise<string>} The job output/trace
  */
-async function getPipelineJobOutput(projectId: string, jobId: number): Promise<string> {
+async function getPipelineJobOutput(projectId: string, jobId: number, limit?: number, offset?: number): Promise<string> {
   projectId = decodeURIComponent(projectId); // Decode project ID
   const url = new URL(
     `${GITLAB_API_URL}/projects/${encodeURIComponent(projectId)}/jobs/${jobId}/trace`
@@ -2632,7 +2634,35 @@ async function getPipelineJobOutput(projectId: string, jobId: number): Promise<s
   }
 
   await handleGitLabError(response);
-  return await response.text();
+  const fullTrace = await response.text();
+  
+  // Apply client-side pagination to limit context window usage
+  if (limit !== undefined || offset !== undefined) {
+    const lines = fullTrace.split('\n');
+    const startOffset = offset || 0;
+    const maxLines = limit || 1000;
+    
+    // Return lines from the end, skipping offset lines and limiting to maxLines
+    const startIndex = Math.max(0, lines.length - startOffset - maxLines);
+    const endIndex = lines.length - startOffset;
+    
+    const selectedLines = lines.slice(startIndex, endIndex);
+    const result = selectedLines.join('\n');
+    
+    // Add metadata about truncation
+    if (startIndex > 0 || endIndex < lines.length) {
+      const totalLines = lines.length;
+      const shownLines = selectedLines.length;
+      const skippedFromStart = startIndex;
+      const skippedFromEnd = startOffset;
+      
+      return `[Log truncated: showing ${shownLines} of ${totalLines} lines, skipped ${skippedFromStart} from start, ${skippedFromEnd} from end]\n\n${result}`;
+    }
+    
+    return result;
+  }
+  
+  return fullTrace;
 }
 
 /**
@@ -3732,8 +3762,8 @@ server.setRequestHandler(CallToolRequestSchema, async request => {
       }
 
       case "get_pipeline_job_output": {
-        const { project_id, job_id } = GetPipelineJobOutputSchema.parse(request.params.arguments);
-        const jobOutput = await getPipelineJobOutput(project_id, job_id);
+        const { project_id, job_id, limit, offset } = GetPipelineJobOutputSchema.parse(request.params.arguments);
+        const jobOutput = await getPipelineJobOutput(project_id, job_id, limit, offset);
         return {
           content: [
             {

package.json

@@ -1,6 +1,6 @@
 {
   "name": "@zereight/mcp-gitlab",
-  "version": "1.0.62",
+  "version": "1.0.63",
   "description": "MCP server for using the GitLab API",
   "license": "MIT",
   "author": "zereight",

schemas.ts

@@ -191,6 +191,8 @@ export const CancelPipelineSchema = z.object({
 export const GetPipelineJobOutputSchema = z.object({
   project_id: z.string().describe("Project ID or URL-encoded path"),
   job_id: z.number().describe("The ID of the job"),
+  limit: z.number().optional().describe("Maximum number of lines to return from the end of the log (default: 1000)"),
+  offset: z.number().optional().describe("Number of lines to skip from the end of the log (default: 0)"),
 });
 
 // User schemas