# Introduction {% embed url="" %} ## Overview Metabob is an advanced static code analysis tool built to run alongside modern generative AI coding assistants—such as Gemini CLI, Copilot CLI, Claude Code, and Cursor. While AI tools accelerate development, Metabob ensures that generated code meets production-grade standards for quality, security, and reliability. Why use Metabob? Read a case study [here.](https://metabob.com/) Running in parallel with AI code generators, Metabob automatically detects issues, including: * Security vulnerabilities * Logic errors * Potential runtime failures * Design flaws * Structural inconsistencies Metabob not only flags defects but also provides targeted annotations and actionable guidance that help AI tools fix issues without introducing regressions. It also recommends follow-up steps, enabling iterative, high-accuracy improvements to your codebase. By integrating Metabob into your development workflow, teams can confidently leverage generative AI while maintaining strong control over code quality, bridging rapid AI generation with rigorous software engineering standards. ## Usage Metabob integrates directly into Agentic AI workflows. While the agents can be instructed to call Metabob autonomously in the AGENTS.md file, users are encouraged to explicitly call Metabob after writing new code or modifying existing files. Metabob provides two primary MCP tools that you can call through the agent to find issues in your codebase: 1. **get\_priority\_issues** \--> Returns a prioritized list of top issues in files the agent has recently worked on **Example prompt to give for an agent:** * *"Use the metabob mcp tool get\_priority\_issues. Only address issues that have not been marked complete or annotated yet."* 2. **search\_codebase\_issues** \--> Searches for issues within a specific category **Example prompts to give for an agent:** * *"Use the metabob\_mcp tool search\_codebase\_issues and look for logic issues. Fix the reported issues that have not been marked complete or annotated yet."* * *"Use the metabob\_mcp tool search\_codebase\_issues and look for security issues. Report the issues to me before proceeding with fixing them."* **Categories available to look for through Metabob:** * Runtime issues * Logic issues * Security issues * Structural issues * Design issues After using Metabob, Metabob and your agent will collaborate and work on your code to fix the detected defects. As a developer, you can sit back and watch it happen. In the background, Metabob provides guidance to your agent that helps it to avoid introducing regressions and prevent the changes from leading you to a fix-and-break loop typical for agentic AI SWE tools. --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://metabob.gitbook.io/product-docs/introduction.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.