Skip to main content
Once you’ve successfully set up Qodo Context Engine, you can start using it in different tools to ask complicated questions about your code.

MCP Usage

Use Qodo Context Engine via MCP.

Qodo Context Engine in Qodo IDE plugin Usage

Use Qodo Context Engine in your IDE.

Qodo Context Engine in Qodo Git plugin Usage

Use Qodo Context Engine in your Git plugin.

​
What can you do with Qodo Context Engine?

​
Get more information about your code

Qodo Context Engine supports a few agents that act as modes for different levels of depth and analysis:
  • Context Ask Agent: Fast, context-aware responses from your indexed codebase.
  • Deep Research Agent: Thorough, multi-repo answers to complex technical queries.
  • Focus Repo Context: Select specific repositories to define scope of context
  • Repo tagging: Tag and group repositories under custom labels

​
What are the core capabilities of Qodo Context Engine?

Context Engine (RAG Indexer) Provides AI agents with accurate, ranked, and scoped context for coding tasks like debugging, refactoring, or explaining unfamiliar logic:
  • Indexes codebases in multiple ways—structural, semantic, embedding-based
  • Creates vector spaces that represent code meaning and relationships
  • Supports high-recall, task-specific retrieval pipelines
Deep Ask An advanced context retrieval agent that performs multi-step reasoning and reflection to solve complex problems by extending the capabilities of standard RAG. Enables the AI to solve more complex tasks that span multiple files, modules, or systems—not just answer short prompts. Context Retriever API and MCP A flexible API layer that exposes Qodo Context Engine’s functionality to internal and external tools, allowing organizations to embed Qodo Context Engine into existing workflows, from developer terminals to CI pipelines and internal tools. Code embedding Improves retrieval precision and makes the system capable of understanding meaning, not just keywords.
  • Converts code into vector form
  • Captures intent, structure, and relationships
  • Powers search, summarization, and reasoning features

​
What are the core use cases?

  • Understand how a feature works across frontend, backend, and data layers
  • Identify the impact of a code change across teams and services
  • Refactor core libraries with full visibility into usage
  • Plan architectural changes with confidence
  • Surface relevant context during code reviews
  • Automatically validate test coverage and suggest improvements
  • Retrieve documentation and usage examples for internal APIs

​
Example usage

​
Comparison based on code behavior

Make informed decisions based on actual implementation rather than just documentation: 
use deep_research: 
Investigate repositories ["langchain-ai/langchain", "BerriAI/litellm"]. 
I don't know which one to use for LLM API calling. Create a comparison and help me decide.
Expected Output: Detailed comparison of both libraries’ implementation approaches, performance characteristics, and suitability for your use case.

​
Research for implementation and planning

Sometimes you need a feature that doesn’t exist in a repository. This example shows how to research and plan a contribution: 
use deep_research:
Investigate repository ["pallets/flask"], is there capability to manage requests queue?
If not, I'd like to submit a PR for the Flask repo to suggest adding a queue for requests.
Therefore, investigate and plan how to do it and create a .md file plan for me to execute.
What this does:
  1. First, the agent verifies if the requested feature already exists
  2. If not, it analyzes the codebase to understand current implementation patterns
  3. Creates a detailed plan aligned with the project’s architecture
  4. Ensures the changes won’t break existing functionality

​
Additional use cases

ScenarioToolExample QueryExpected Outcome
🏛️ Understanding system architecturedeep_research”Explain how our microservices communicate and what protocols they use”Detailed explanation of service communication patterns, protocols, and data flow
🎨 Finding design patternsget_context”singleton pattern implementation”Code examples of singleton patterns used in the codebase
🚨 Locating error handling patternsget_context”try catch error handling with logging”Examples of error handling patterns with logging
💰 Understanding business logicdeep_research”How is pricing calculated for premium users?”Detailed explanation of pricing logic and rules
🔐 Analyzing authentication flowdeep_research”Trace the complete OAuth2 authentication flow”Step-by-step authentication process across services
⚠️ Identifying security vulnerabilitiesissues[code diff with auth changes]Potential security issues in authentication changes
⚡ Planning feature additionsdeep_research”Where should we add caching for better performance?”Strategic caching recommendations
🔥 Understanding error sourcesdeep_research”What could cause a 500 error in the checkout process?”Potential failure points and error conditions
🔌 Planning third-party integrationsget_context”stripe payment integration”Existing integration patterns and implementations
✅ Validating best practicesdeep_research”Are we following REST best practices in our API design?”Analysis of REST compliance and recommendations