CodeBin

    1

    High-performance binary symbol indexer for $O(1)$ codebase lookups and context-efficient navigation.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    CodeBin

    Example session with this skill installed

    Find the signature and description for the 'calculate_tax' function in the billing module.

    • Read your context and instructions
    • Compiled the codebin

    I have retrieved the symbol details for calculate_tax from the CodeBin binary index.

    Symbol Lookup Result

    FieldValue
    Namesrc/billing.py::calculate_tax
    TypeFunction (F)
    StatusActive
    Signaturedef calculate_tax(gross_income, state)
    Locationsrc/billing.py (Line 42)
    DescriptionCalculates regional tax brackets based on annual gross income.
    Last Modified1730250000000

    Next steps

    • Review the src/billing.py file if you need to see the internal logic of the tax calculation.
    • Search for other symbols in billing.py to see related helper functions.
    • Run a new scan if you have recently modified the tax logic to update the index.

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

    About this skill

    Stop wasting context window tokens reading entire source files just to find a function signature. CodeBin indexes your codebase into a high-performance binary repository, allowing agents to perform $O(1)$ symbol lookups in under 1ms.

    What it does

    • Instant Lookup retrieves signatures, types, and line numbers without reading raw files.
    • Incremental Scanning uses AST content hashes to update the index only when files change.
    • Multi-token Search filters codebase symbols by keyword and type (Class, Function, Method).
    • Tombstone Tracking records symbol removals and deprecations with historical timestamps.
    • Binary Packing stores data in a compact TLV format to minimize disk I/O and latency.

    How it works

    1. Initialize the .codebin directory using the engine to set up the binary store.
    2. Scan the project to parse symbols into the indexed binary repository.
    3. Lookup specific symbols via qualified paths like src/auth.py::login for immediate retrieval.
    4. Search the index using keywords to find relevant classes or functions across the repository.

    Frameworks & tools

    Built with Python using standard library ast for static extraction. Supports Model Context Protocol (MCP) for integration with Cursor, Windsurf, and Claude Desktop.

    Why this beats prompting it yourself

    Manual file reading consumes massive amounts of context tokens and increases latency. CodeBin provides sub-millisecond access to precise symbol data, keeping your agent's context clean and focused on logic rather than navigation.

    Use cases

    • Check function parameters and return types without opening the source file.
    • Track codebase changes and symbol removals across commits.
    • Rapidly onboard an agent to a large repository by indexing symbols upfront.
    • Audit class inheritance and variable definitions in dense projects.

    Known limitations

    Extraction is limited to static AST analysis. It does not support dynamic attributes generated at runtime or symbols requiring code execution.

    Only supports static AST analysis. No support for dynamic runtime attributes or symbols requiring code execution. Requires Python environment for scripts.

    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

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 10 days ago

    • Free with an Agensi account

    Requires

    Python
    Model Context Protocol Setup MCP
    macOS
    Git

    Permissions

    Terminal / Shell
    Write Files
    Network Access
    Read Files
    Listed10 days ago

    What's inside

    Frequently Asked Questions

    CodeBin solves 'context bloat' by indexing your code into a compact binary format, allowing your AI agent to fetch specific function signatures and types instantly without wasting thousands of tokens reading entire files.

    CodeBin is fully compatible with any tool supporting the Model Context Protocol (MCP), including Cursor, Windsurf, and Claude Desktop, as well as standalone Python environments.

    No, CodeBin uses incremental scanning with AST content hashes, meaning it only re-indexes the specific files you have modified, keeping the overhead minimal even in large repositories.

    Currently, CodeBin uses static AST analysis which is highly optimized for Python; however, support for other languages via tree-sitter is planned for future updates.

    Setup is straightforward: once installed, you simply run the initialization command in your project root to create a `.codebin` directory, then use the scan command to build your first index.

    CodeBin uses static analysis only; it does not execute your code or handle dynamic attributes generated at runtime, ensuring a secure and predictable indexing process.

    New in LLM Ops