feat(mcp): add style parameter support to MCP tools#839
Conversation
Add support for specifying output format style (xml, markdown, json, plain) in MCP pack tools and attach tool. This allows users to choose the most appropriate format for their AI analysis workflows. Changes: - Add style parameter to packCodebaseTool with detailed format descriptions - Add style parameter to packRemoteRepositoryTool with detailed format descriptions - Update attachPackedOutputTool to handle multiple input formats (xml, md, txt, json) - Add format-specific parsing functions for each output style - Update tool descriptions to explain the differences between formats - Fix tests to handle new function signatures and add path.extname mock The style parameter defaults to 'xml' for backward compatibility.
|
Important Review skippedAuto incremental reviews are disabled on this repository. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the Note Other AI code review bot(s) detectedCodeRabbit has detected other AI code review bot(s) in this pull request and will avoid duplicating their findings in the review comments. This may lead to a less comprehensive review. WalkthroughAdds multi-format support (XML, Markdown, JSON, plain text) across packing and attaching tools. Introduces style-aware output selection and dynamic filenames. Implements format detection and parsing in attachPackedOutputTool, with format-specific extractors and processedFiles generation. Updates input schemas and handler signatures to include a style option. Adjusts tests for unsupported extensions. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Caller
participant AttachTool as attachPackedOutputTool
participant FS as FileSystem
participant Parser as Format Extractors
Caller->>AttachTool: call({ path })
AttachTool->>AttachTool: resolvePathAndFormat(path)
alt Path is directory
AttachTool->>FS: check defaultFilePathMap[style] candidates
FS-->>AttachTool: first existing file path + inferred format
else Path is file
AttachTool->>AttachTool: detect format by extension
end
AttachTool->>FS: readFile(resolvedPath)
FS-->>AttachTool: content
AttachTool->>Parser: extractFileMetrics(content, format)
Parser-->>AttachTool: { filePaths, charCounts }
AttachTool->>AttachTool: compute totals, tokens, processedFiles
AttachTool-->>Caller: formatPackToolResponse({...}) or error JSON
sequenceDiagram
autonumber
actor Caller
participant PackCodebase as pack_codebase
participant CLI as repomix CLI
participant FS as FileSystem
Caller->>PackCodebase: call({ directory, style?, ... })
PackCodebase->>PackCodebase: style := style || 'xml'
PackCodebase->>CLI: run({ directory, style, ... })
CLI-->>PackCodebase: exit code / result
PackCodebase->>PackCodebase: outputPath := join(directory, defaultFilePathMap[style])
PackCodebase-->>Caller: formatPackToolResponse(outputPath)
sequenceDiagram
autonumber
actor Caller
participant PackRemote as pack_remote_repository
participant CLI as repomix CLI
participant FS as FileSystem
Caller->>PackRemote: call({ remote, style?, ... })
PackRemote->>PackRemote: style := style || 'xml'
PackRemote->>CLI: run({ remote, style, ... })
CLI-->>PackRemote: exit code / result
PackRemote->>PackRemote: outputPath := join(workdir, defaultFilePathMap[style])
PackRemote-->>Caller: formatPackToolResponse(outputPath)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Possibly related PRs
Pre-merge checks and finishing touches✅ Passed checks (3 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![gemini-code-assist[bot]](https://avatars.githubusercontent.com/in/956858?s=60&v=4){kind=small}
There was a problem hiding this comment.
Summary of Changes
Hello @yamadashy, I'm Gemini Code Assist1! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!
This pull request significantly enhances the flexibility of MCP tools by introducing support for multiple output formats (XML, Markdown, JSON, and Plain text) when packing codebases or remote repositories. The packCodebaseTool and packRemoteRepositoryTool now include a style parameter, allowing users to specify their preferred output format. Concurrently, the attachPackedOutputTool has been upgraded to automatically detect and correctly parse content from any of these new formats, streamlining AI analysis workflows. The changes maintain backward compatibility by defaulting to XML and include updated tool descriptions for clarity.
Highlights
- Multi-format Output Support: The
packCodebaseToolandpackRemoteRepositoryToolnow accept astyleparameter, enabling output in XML, Markdown, JSON, or Plain text formats. - Enhanced Input Parsing: The
attachPackedOutputToolhas been updated to intelligently detect and parse these new output formats, improving flexibility for AI analysis. - Backward Compatibility: XML remains the default output style for all packing tools, ensuring existing workflows are unaffected by these changes.
- Comprehensive Format Descriptions: Tool descriptions now include detailed explanations of each supported output format to guide users in choosing the appropriate style.
- Updated Test Coverage: Tests for
attachPackedOutputToolhave been expanded to validate the new multi-format parsing capabilities and error handling for unsupported formats.
Using Gemini Code Assist
The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.
Invoking Gemini
You can request assistance from Gemini at any point in your pull request via creating an issue comment (i.e. comment on the pull request page) using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands.
| Feature | Command | Description |
|---|---|---|
| Code Review | /gemini review |
Performs a code review for the current pull request in its current state. |
| Pull Request Summary | /gemini summary |
Provides a summary of the current pull request in its current state. |
| Comment | @gemini-code-assist | Responds in comments when explicitly tagged, both in issue comments and review comments. |
| Help | /gemini help |
Displays a list of available commands. |
Customization
To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.
Limitations & Feedback
Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.
You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.
Footnotes
-
Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩
There was a problem hiding this comment.
Pull Request Overview
This PR adds support for specifying output format styles in MCP tools, allowing users to choose between XML, Markdown, JSON, and Plain text formats for their AI analysis workflows. This enhances the flexibility of the MCP tools while maintaining backward compatibility with XML as the default format.
- Added style parameter to packCodebaseTool and packRemoteRepositoryTool with format descriptions
- Enhanced attachPackedOutputTool to handle multiple input formats (xml, md, txt, json) with format-specific parsing functions
- Updated tests to handle new function signatures and mock additional path utilities
Reviewed Changes
Copilot reviewed 4 out of 4 changed files in this pull request and generated 2 comments.
| File | Description |
|---|---|
| src/mcp/tools/packCodebaseTool.ts | Added style parameter with detailed format descriptions and dynamic output filename |
| src/mcp/tools/packRemoteRepositoryTool.ts | Added style parameter with format descriptions and output filename mapping |
| src/mcp/tools/attachPackedOutputTool.ts | Enhanced to support multiple formats with format detection and parsing functions |
| tests/mcp/tools/attachPackedOutputTool.test.ts | Updated test mocks and test case description for new functionality |
| while (true) { | ||
| match = fileRegex.exec(content); | ||
| if (!match) { | ||
| break; | ||
| } |
There was a problem hiding this comment.
Replace the infinite loop with a more idiomatic while condition. Use while ((match = fileRegex.exec(content)) !== null) instead of the infinite loop with break statement.
| while (true) { | ||
| match = fileRegex.exec(content); | ||
| if (!match) { | ||
| break; | ||
| } |
There was a problem hiding this comment.
Replace the infinite loop with a more idiomatic while condition. Use while ((match = fileRegex.exec(content)) !== null) instead of the infinite loop with break statement.
Deploying repomix with
|
| Latest commit: |
3f86488
|
| Status: | ✅ Deploy successful! |
| Preview URL: | https://1f3304b9.repomix.pages.dev |
| Branch Preview URL: | https://feat-mcp-style.repomix.pages.dev |
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## main #839 +/- ##
==========================================
+ Coverage 87.13% 87.27% +0.14%
==========================================
Files 113 113
Lines 6723 6839 +116
Branches 1384 1408 +24
==========================================
+ Hits 5858 5969 +111
- Misses 865 870 +5 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
There was a problem hiding this comment.
Code Review
This pull request effectively introduces support for multiple output formats (XML, Markdown, JSON, Plain) in the MCP tools. The implementation is generally solid, with good separation of concerns for parsing different formats. My review includes suggestions for improving code maintainability by using more modern JavaScript features and reducing code duplication. The most critical feedback is the absence of unit tests for the new parsing logic for Markdown, JSON, and plain text formats. It's crucial to add these tests to ensure the new functionality is robust and reliable.
|
|
||
| test('should handle non-XML file error', async () => { | ||
| const testFilePath = '/test/not-xml-file.txt'; | ||
| test('should handle non-supported file format error', async () => { |
There was a problem hiding this comment.
This pull request introduces significant new parsing logic for Markdown, Plain text, and JSON formats, but there are no tests covering these new code paths. Please add unit tests for extractFileMetricsMarkdown, extractFileMetricsPlain, and extractFileMetricsJson. You should also add integration tests within this file that exercise the attachPackedOutputTool with .md, .txt, and .json files to ensure they are resolved and parsed correctly.
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 4
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tests/mcp/tools/attachPackedOutputTool.test.ts (1)
77-107: Add tests for markdown/json/plain and CRLF parsing.Current tests only cover XML. Add positive cases for each format and a CRLF variant to guard regexes.
I can add fixtures and tests for:
- Markdown: “## File: …” + fenced code
- Plain: “================” separators
- JSON:
{ "files": { "path": "content" } }and array variant- CRLF versions of each
Also applies to: 180-192
🧹 Nitpick comments (8)
src/mcp/tools/packRemoteRepositoryTool.ts (2)
39-43: Validate bounds for topFilesLength.Permit only reasonable integers (e.g., >=1). Currently any number passes, including negatives/NaN.
- topFilesLength: z - .number() + topFilesLength: z + .number().int().min(1) .optional() .default(10)
44-49: Schema nit: default already implies optional.
.default('xml')makes the field optional;.optional()is redundant.- style: repomixOutputStyleSchema - .optional() - .default('xml') + style: repomixOutputStyleSchema.default('xml')src/mcp/tools/packCodebaseTool.ts (2)
35-39: Validate bounds for topFilesLength.Mirror the remote tool: enforce integer >=1.
- topFilesLength: z - .number() + topFilesLength: z + .number().int().min(1) .optional() .default(10)
40-45: Schema nit: default already implies optional.- style: repomixOutputStyleSchema - .optional() - .default('xml') + style: repomixOutputStyleSchema.default('xml')src/mcp/tools/attachPackedOutputTool.ts (4)
164-182: XML entity decoding for accurate char counts.If the XML encoder escapes characters, lengths are inflated. Decode entities before counting.
-const fileContent = match[2]; +let fileContent = match[2]; +fileContent = fileContent + .replaceAll('<', '<') + .replaceAll('>', '>') + .replaceAll('&', '&') + .replaceAll('"', '"') + .replaceAll(''', "'");
187-207: Windows line endings support.Use
\r?\nin regex to parse CRLF files.-const fileRegex = /## File: ([^\n]+)\n```[^\n]*\n([\s\S]*?)```/g; +const fileRegex = /## File: ([^\r\n]+)\r?\n```[^\r\n]*\r?\n([\s\S]*?)```/g;-const fileRegex = /={16,}\nFile: ([^\n]+)\n={16,}\n([\s\S]*?)(?=\n={16,}\n|$)/g; +const fileRegex = /={16,}\r?\nFile: ([^\r\n]+)\r?\n={16,}\r?\n([\s\S]*?)(?=\r?\n={16,}\r?\n|$)/g;Also applies to: 212-232
241-254: JSON variant: support array-form as well.In case JSON packs use
{ files: [{ path, content }] }, add a fallback.-const files = jsonData.files || {}; - -for (const [filePath, fileContent] of Object.entries(files)) { +const files = jsonData.files || {}; +if (Array.isArray(files)) { + for (const f of files) { + if (f && typeof f.path === 'string' && typeof f.content === 'string') { + filePaths.push(f.path); + fileCharCounts[f.path] = f.content.length; + fileContents[f.path] = f.content; + } + } +} else { + for (const [filePath, fileContent] of Object.entries(files)) { if (typeof fileContent === 'string') { filePaths.push(filePath); fileCharCounts[filePath] = fileContent.length; + fileContents[filePath] = fileContent; } } +}
286-291: Reduce noisy console.warn in library code.Prefer a debug logger or remove to keep MCP logs clean.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (4)
src/mcp/tools/attachPackedOutputTool.ts(7 hunks)src/mcp/tools/packCodebaseTool.ts(4 hunks)src/mcp/tools/packRemoteRepositoryTool.ts(4 hunks)tests/mcp/tools/attachPackedOutputTool.test.ts(2 hunks)
🧰 Additional context used
🧬 Code graph analysis (3)
src/mcp/tools/attachPackedOutputTool.ts (2)
src/config/configSchema.ts (1)
defaultFilePathMap(9-14)src/core/file/fileTypes.ts (1)
ProcessedFile(6-9)
src/mcp/tools/packRemoteRepositoryTool.ts (2)
src/config/configSchema.ts (2)
repomixOutputStyleSchema(5-5)defaultFilePathMap(9-14)src/mcp/tools/mcpToolRuntime.ts (1)
createToolWorkspace(49-59)
src/mcp/tools/packCodebaseTool.ts (2)
src/config/configSchema.ts (2)
repomixOutputStyleSchema(5-5)defaultFilePathMap(9-14)src/mcp/tools/mcpToolRuntime.ts (1)
createToolWorkspace(49-59)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (4)
- GitHub Check: Build and run with Bun (windows-latest, latest)
- GitHub Check: Test (windows-latest, 24.x)
- GitHub Check: Test (windows-latest, 20.x)
- GitHub Check: Cloudflare Pages
🔇 Additional comments (4)
tests/mcp/tools/attachPackedOutputTool.test.ts (3)
48-51: Good: extname mock ensures extension routing is testable.
149-156: Nice rename; broadened unsupported extension test matches new formats.
69-75: Tool registration assertions look good.src/mcp/tools/packCodebaseTool.ts (1)
86-88: Drop redundant 'xml' fallbacks; rely on input-schema default for styleThe Zod input schemas default style to 'xml', so remove "|| 'xml'" and pass style directly (no type-change needed). Apply to:
- src/mcp/tools/packCodebaseTool.ts (≈ lines 86–98)
- src/mcp/tools/packRemoteRepositoryTool.ts (≈ lines 83–96)
- Replace while loops with idiomatic for...of matchAll() patterns in all extractFileMetrics functions - Generate supportedExtensions dynamically from defaultFilePathMap instead of hardcoding - Remove console.warn from JSON parser for consistency with other format parsers - Add comprehensive tests for new Markdown, Plain, and JSON parser functions Addresses all automated review feedback from CodeRabbit, Copilot, and Gemini Code Assist.
- Remove unused match variables from extractFileMetrics functions - Fix unused error parameter in JSON parser catch block - Correct test expectations for Markdown parsing (account for newlines in extracted content) - Ensure all tests properly declare result variables where needed All tests pass and lint issues resolved.
- Add CRLF support for Windows line endings in Markdown and Plain parsers - Improve Zod schema validation: topFilesLength now requires integer >= 1 - Remove redundant || 'xml' fallbacks as Zod defaults handle this - Clean up redundant .optional() in schema definitions with .default() - Add comprehensive CRLF tests for Markdown and Plain text formats These changes improve cross-platform compatibility and input validation.
|
/gemini review |
There was a problem hiding this comment.
Code Review
This pull request introduces support for multiple output formats (XML, Markdown, JSON, Plain text) for the MCP tools, which is a great enhancement for flexibility. The changes are well-structured, with new parameters added to the packing tools and corresponding parsing logic in the attachPackedOutputTool. The implementation includes format detection, dedicated parsing functions for each style, and updated tests to ensure correctness. I've identified a couple of areas for improvement in attachPackedOutputTool.ts to enhance maintainability and robustness, but overall this is a solid contribution.
This PR adds support for specifying output format styles in MCP tools, allowing users to choose between XML, Markdown, JSON, and Plain text formats for their AI analysis workflows.
Summary
styleparameter topackCodebaseToolandpackRemoteRepositoryToolattachPackedOutputToolto handle multiple input formats (xml, md, txt, json)<file>tags (default)##headers and code blocksChanges
Modified Files
src/mcp/tools/packCodebaseTool.ts- Added style parameter with detailed descriptionssrc/mcp/tools/packRemoteRepositoryTool.ts- Added style parameter with detailed descriptionssrc/mcp/tools/attachPackedOutputTool.ts- Added multi-format support and parsing functionstests/mcp/tools/attachPackedOutputTool.test.ts- Updated tests for new functionalityNew Features
Checklist
npm run testnpm run lint