More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    🧰 Tool Schema Linter

    2

    The Tool Schema Linter is a specialized audit tool for developers building agentic workflows and tool-calling applications.

    $12

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    🧰 Tool Schema Linter

    🧰 Tool Schema Linter

    Example session with this skill installed

    Lint my function definitions in ./src/tools/weather.json and check if they are well-defined.

    • Read your context and instructions
    • Compiled the tool schema linter

    Found 3 issues in weather.json:

    • [TSL002] Severity: High | Param 'unit' in 'get_weather' missing description.
    • [TSL008] Severity: Med | Param 'location' is free-form; recommend adding 'examples' or boundaries.
    • [TSL004] Severity: Low | 'get_temp' and 'get_weather' have overlapping purposes.
    Found 3 issues in weather.json

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

    About this skill

    What it does

    The Tool Schema Linter is a specialized audit tool for developers building agentic workflows and tool-calling applications. It analyzes your JSON function definitions against a rigorous set of rules designed to prevent LLM hallucinations, "lazy" tool selection, and security risks. By scanning your schema files, it identifies missing descriptions, untyped parameters, and overlapping tool signatures that often cause agents to fail in production.

    Why use this skill

    Writing perfect function schemas for LLMs is tedious and error-prone. While a basic prompt might catch a missing comma, this skill uses a structured heuristic engine to find logical flaws, such as too many tools for a single context window or free-form parameters that should be bounded to an enum for reliability. It saves hours of debugging "why the agent didn't call the right tool" by ensuring your definitions are mathematically and semantically sound.

    Supported tools

    • Standard JSON function-calling schemas (OpenAI, Anthropic, Gemini formats)
    • Local tool definition files (.json, .tool.json)
    • Standard input streams for CI/CD integration
    • Customizable rule thresholds via reference JSON files

    Heuristic detector. It checks the schema you give it; the unsafe-tool rule is advisory, and overlap detection works from tool names and purposes, so it flags candidates for review rather than asserting a duplicate.

    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 5 months ago

    Permissions

    Read Files
    Terminal / Shell

    File Scopes

    **/tools.json
    **/*tools*.json
    **/functions*.json
    **/*.tool.json
    --stdin

    Read-only. The too-many-tools threshold is configurable. Reads no environment variables and writes nothing.

    Listed5 months ago

    What's inside

    Frequently Asked Questions

    This skill helps prevent LLM 'hallucinations' and incorrect tool calls by scanning your function definitions for missing descriptions, untyped parameters, and logical overlaps that confuse agents.

    Yes, the linter is compatible with standard JSON function-calling formats used by OpenAI, Anthropic, and Google Gemini, as well as local .json tool definition files.

    Unlike a standard JSON validator that only checks for syntax errors, this tool performs a semantic audit to flag issues like near-duplicate tools, missing safety guards, and parameters that should be restricted to enums for better reliability.

    The skill is designed for easy integration into your development workflow; you can run it via standard input streams, making it an ideal step for your CI/CD pipelines to catch schema errors before deployment.

    The linter will issue a warning if the number of tools in your schema exceeds a threshold known to degrade model selection accuracy, helping you maintain a high success rate for tool calls.

    You receive the core Tool Schema Linter engine, documentation for implementation, and the ability to customize rule thresholds via reference files to suit your specific project needs.

    New in Agents & Orchestration