Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    graf intelligence

    4

    Turn AI coding agents into codebase intelligence assistants using Graf's local-first knowledge graph and MCP server.

    $4.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    graf intelligence

    graf intelligence

    Example session with this skill installed

    Use Graf Intelligence to analyse my current project.

    Connect to Graf's MCP server and inspect the indexed codebase.

    I want a complete technical audit covering:

    1. Project architecture and structure.
    2. Main modules and their responsibilities.
    3. Important functions and their relationships.
    4. Dependency analysis.
    5. Potential architectural problems.
    6. Areas that may need refactoring.

    Use actual Graf MCP tools and provide evidence from the indexed project. Do not invent files, functions or dependencies.

    Finish with a structured technical report and recommended improvements.

    • Read your context and instructions
    • Compiled the graf intelligence

    Graf Intelligence: Project Audit

    1. Executive Summary

    Graf Intelligence analysed the indexed project using Graf's available codebase intelligence tools.

    The report covers project structure, architecture, code relationships, dependencies and potential areas for improvement.

    1. Project Architecture

    Area| Findings
    Project structure| Derived from indexed files
    Main modules| Identified through code entities
    Dependencies| Retrieved from Graf relationships
    Entry points| Identified where available

    1. Code Relationships

    The analysis examines

    • Important functions and their callers.
    • Connections between modules.
    • Dependencies between code entities.
    • Related components and implementation paths.
    1. Potential Issues

    Each finding should include

    • Issue description.
    • Relevant file and code location.
    • Supporting Graf context.
    • Potential impact.
    • Recommended improvement.

    Only verified findings should be presented as confirmed issues.

    1. Recommended Improvements

    Prioritise improvements based on

    1. Correctness and reliability.

    2. Architectural maintainability.

    3. Dependency complexity.

    4. Code organisation.

    5. Documentation quality.

    6. Final Assessment

    Graf Intelligence provides a structured view of the project using Graf's indexed codebase context.

    The report distinguishes verified findings from assumptions and identifies areas requiring additional investigation.

    Next Steps

    • Review the identified architectural concerns.
    • Investigate high-impact dependencies.
    • Validate proposed refactoring.
    • Re-run analysis after significant changes.

    This is an example report structure. Actual findings and metrics must come from the connected Graf installation and the target repository.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    Graf Intelligence

    Give your AI coding agent a deeper understanding of your software projects with Graf.

    Graf Intelligence is an AI agent skill that connects compatible coding agents to Graf's local-first codebase intelligence workspace through its Model Context Protocol (MCP) server.

    Instead of relying entirely on basic file searches, your agent can use Graf's indexed codebase context and relationships to investigate, understand and navigate complex software projects.

    What it can do

    • Codebase Exploration: Understand project structure, code entities and relationships.
    • Dependency Analysis: Explore connections between functions, modules and components.
    • Bug Investigation: Investigate problems using indexed code and contextual relationships.
    • Architecture Understanding: Generate structured explanations of software architecture.
    • Change Impact Analysis: Investigate potentially affected components before making changes.
    • Git Intelligence: Use available project history and branch information.
    • Project Documentation: Generate technical reports and documentation using verified project context.

    How it works

    Graf Intelligence provides structured instructions and workflows for compatible AI coding agents.

    It uses Graf's existing MCP server to access available codebase intelligence capabilities.

    Requirements

    • Graf installed locally.
    • Supported Node.js runtime.
    • Graf MCP server configured in your AI coding agent.
    • An indexed project.

    Compatibility

    Designed for compatible AI coding agents, including Claude Code, Cursor and other agents supporting the required Agent Skills and MCP integrations.

    Built for developers

    Created by Gurjotpal Singh, the developer behind Graf.

    Website: https://gurjotpalsingh.in

    GitHub: https://github.com/SinghGurjot01-Code

    • Requires Graf to be installed separately. The skill does not include the Graf application or MCP server.
    • Requires Node.js 22.5.0 or later for the Graf runtime.
    • The Graf MCP server must be configured and running in the buyer's AI coding agent.
    • Analysis is limited to files and code structures supported by Graf's indexing and parsing engine.
    • Supported file types include JavaScript, TypeScript, JSX, TSX, HTML, CSS, SCSS, Vue, Svelte, Python, Java, C#, YAML and JSON.
    • Files containing syntax errors may fail to index correctly and can produce incomplete analysis.
    • Newly created or modified files may require synchronization before their latest contents appear in the knowledge graph.
    • Git must be installed for Git branch and commit history features.
    • macOS and Linux currently use the CLI installation workflow; the dedicated desktop installer is provided for Windows.
    • AI-generated analysis depends on the connected AI agent and the quality and freshness of the indexed project data.
    • Graf memory entries proposed by AI agents require human approval before becoming active.
    • Large projects may experience SQLite indexing or query latency depending on hardware and project complexity.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 8 days ago

    Requires

    Windows
    macOS
    Node.js
    Git

    Permissions

    Read Files
    Write Files
    Terminal / Shell
    Network Access
    Browser

    File Scopes

    **

    Graf Intelligence requires access to the customer's project files through their configured Graf MCP server. File access is used for codebase exploration, dependency analysis and software architecture understanding. No external account connections or environment variable access are required. The skill does not require permission to modify project files.

    Listed8 days ago
    Updated4 days ago

    What's inside

    Frequently Asked Questions

    Unlike generic prompts that require uploading entire files, this skill uses a local SQLite knowledge graph to retrieve only relevant AST nodes, reducing token usage by over 70% and preventing context window overflow in large repositories.

    This skill is built on the Model Context Protocol (MCP), making it compatible with any AI agent or IDE that supports MCP, such as Claude Desktop or various developer-focused AI extensions.

    Graf Intelligence currently offers deep integration for JavaScript, HTML, and CSS, utilizing Git version control systems to track changes and analyze the impact of refactors across these languages.

    No, the skill operates entirely within your local environment using a SQLite database; it does not make outbound network requests, ensuring your proprietary source code remains private.

    The 'impact_of' tool performs a blast radius analysis, mapping every downstream function, component, or DOM element affected by a specific change so you can identify potential breakages before they happen.

    Your purchase includes the Graf Intelligence MCP server, the SQLite indexing engine, and a suite of specialized tools for tree mapping, symbol searching, and dependency analysis.

    New in LLM Ops