deepcontext-mcp

🥈Silver

DeepContext adds symbol-aware semantic search to Claude Code, Codex CLI, and other agents. It enables faster, smarter context retrieval for large codebases. Developers and operations teams benefit from improved code navigation and understanding.

274240Updated Yesterday

Intermediate30min to implementautomation

Saves ~210 min per use

Quick InstallView Source

git clone https://github.com/Wildcard-Official/deepcontext-mcp.git

Works with:

Claude

Overview

About This Skill

How to Use

["Install DeepContext-MCP in your environment (Claude Code or Codex CLI). Run `mcp install deepcontext` or follow setup instructions for your IDE.","Identify the symbol you want to analyze (class name, function name, variable name, etc.) and the project context. Replace [SYMBOL_NAME] and [PROJECT_NAME] in the prompt template.","Specify the context area you care about (e.g., 'database layer', 'API endpoints', 'UI components') to focus the search. Use [CONTEXT_AREA] placeholder.","Run the analysis and review the output. Use the prioritized list of files to plan your refactoring or documentation efforts.","For large codebases, run DeepContext-MCP incrementally by breaking down the analysis into smaller modules or layers. Use the `--limit` flag if available to scope the search."]

Setup & Installation

Quick Install

No install command available. Check the GitHub repository for manual installation instructions.

Alternative Install (Git Clone)

git clone https://github.com/Wildcard-Official/deepcontext-mcp

Requirements

Claude Code or compatible AI agent
Works with: Claude

Quick Start Guide

Install the Skill

Copy the install command above and run it in your terminal.

Open Your AI Agent

Launch Claude Code, Cursor, or your preferred AI coding agent.

Try It Out

Use the prompt template or examples below to test the skill.

Customize

Adapt the skill to your specific use case and workflow.

Usage Examples

Prompt Template

Use DeepContext-MCP to analyze the [PROJECT_NAME] codebase. Find all references to the [SYMBOL_NAME] symbol across all files. Summarize how [SYMBOL_NAME] is used in [CONTEXT_AREA] (e.g., API layer, database layer, UI layer). Highlight any inconsistencies or deprecated patterns in its usage. Provide a prioritized list of files that need updates based on this analysis.

Example Output

DeepContext-MCP analysis of the `auth-service` codebase reveals the following insights about the `JWTValidator` symbol:

**Cross-Referenced Usage:**
- Found in 24 files across 6 modules (auth-api, auth-models, auth-utils, auth-tests, auth-integration, auth-docs)
- Primary usage in `auth-api/src/handlers/token_handler.py` (34 occurrences) and `auth-utils/src/jwt_utils.py` (18 occurrences)

**Contextual Breakdown:**
- **API Layer (6 files):** Used in request validation middleware and token refresh endpoints. All instances follow consistent pattern with `JWTValidator.validate(token)` calls.
- **Database Layer (4 files):** Only referenced in `auth-models/src/user_repository.py` for token blacklisting. Implementation uses deprecated `JWTValidator.blacklist()` method (marked for removal in v2.3).
- **UI Layer (2 files):** Found in `auth-frontend/src/components/LoginForm.tsx` for client-side validation. Uses outdated `JWTValidator.decode()` which doesn't verify expiration.

**Critical Findings:**
1. **Deprecated API Usage:** 12 files still use `JWTValidator.blacklist()` which will break in next release. Priority: HIGH
2. **Inconsistent Validation:** Client-side validation in `LoginForm.tsx` doesn't check token expiration, creating security risk. Priority: CRITICAL
3. **Missing Documentation:** No usage examples in `auth-docs/` despite being core component. Priority: MEDIUM

**Recommended Actions:**
1. Update all `blacklist()` calls to new `TokenBlacklistService` (see migration guide in auth-docs)
2. Refactor `LoginForm.tsx` to use server-side validation endpoint
3. Add symbol usage examples to documentation