Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+22 more

    hex local

    1

    Query version-accurate Hex documentation and GitHub issues directly within your AI coding agent.

    $3

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    hex local

    hex local

    Example session with this skill installed

    The project uses Req 0.7.0. How do I merge options into a request? Check the docs for the exact syntax and give me an example.

    • Read your context and instructions
    • Compiled the hex local

    I have retrieved the documentation for req version 0.7.0 using the hex_local index.

    Req.merge/2 Documentation

    In req version 0.7.0, the Req.merge/2 function is used to merge options into a request struct. This is particularly useful when you have a base request and need to add headers, params, or other options dynamically.

    AttributeValue
    Packagereq
    Version0.7.0
    FunctionReq.merge(request, options)
    Returns%Req.Request{}
    DocumentationReq 0.7.0 - Req.merge/2
    Sourcereq.ex at v0.7.0, line 551

    Implementation Example

    
    # Create a base request
    request =
      Req.new(
        base_url: "https://httpbin.org",
        params: [page: 1, limit: 20]
      )
    
    # Merge additional headers and params
    request =
      Req.merge(request,
        params: [page: 2],
        headers: [{"accept", "application/json"}]
      )
    
    # The resulting request now contains both the base_url and the new options
    request.options[:params]
    #=> [limit: 20, page: 2]
    

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

    About this skill

    When coding Elixir or Erlang, agents often hallucinate library APIs or suggest deprecated patterns from outdated training data. This skill guides an AI agent to query Hex documentation matched to your project’s dependencies and investigate relevant upstream GitHub issues.. It ensures the agent works with the actual code in your environment, not just what it remembers from its training cutoff.

    Requirements

    A separately installed and connected Hex Local MCP server with a configured embedding model. Hosted providers require an API key; local providers such as Ollama do not. Chat models are only required for optional memory curation

    What it does

    • Version-matched retrieval pulls documentation for the exact package versions defined in your project's lockfile.
    • Hex library discovery searches for packages based on capabilities or keywords when the specific dependency name is unknown.
    • Deep documentation search performs hybrid keyword and semantic lookups across API references, guides, and changelogs.
    • Contextual extraction retrieves surrounding documentation chunks to provide surrounding context from the same documentation entry of a function or macro.
    • Upstream issue tracking checks GitHub for bug reports or known behaviour issues when local code acts unexpectedly.

    How it works

    1. With PROJECT_ROOT configured, the MCP server reads your project’s mix.lock to resolve package versions. Otherwise, the agent supplies the locked version explicitly.
    2. It uses search_docs to find relevant API examples, typespecs, or configuration guides within the hex_local MCP server.
    3. If information is missing or partial, the agent uses get_doc to pull neighboring context or follows next_id links.
    4. The agent cross-references findings with upstream GitHub issues if it suspects a dependency bug.

    Frameworks & tools

    Specifically built for the Elixir and Erlang ecosystem. It interacts with mix, HexDocs, and GitHub. Requires a functional hex_local MCP server with a configured embedding provider like Ollama or a hosted API.

    Why this beats prompting it yourself

    Generic prompts can produce plausible but outdated API advice. This skill directs the agent to version-matched documentation and upstream evidence, with citations that make its conclusions easier to verify.

    Use cases

    • Troubleshooting unexpected behavior in a specific version of a dependency.
    • Implementing complex macro-generated functions that aren't visible in raw source code.
    • Finding the right NATS or Redis client by searching for capabilities rather than names.
    • Porting code between different versions of Phoenix or Ecto using precise changelogs.

    Known limitations

    Requires a separately installed Hex Local MCP server. Only one version per package is retained in the local index at a time. Retrieval quality depends on the chosen embedding provider. GitHub issues may describe other versions or unreleased changes, so the agent must verify their applicability. Cached documentation may be retained with a version notice

    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 passes our trust checks below.

    Security scanned

    Passed 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 scanned

    Verified clean today

    • Passed all security checks, Safe to install

    Requires

    OpenAI
    Anthropic
    Google AI
    Hex Local MCP

    Listedtoday

    What's inside

    Frequently Asked Questions