More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Context Cost Audit — what does your Claude Code pack cost?

    1

    Measures what a CLAUDE.md, rules folder or skill/agent pack costs in context before your first prompt (startup) versus only when triggered (on-demand), then tells you what to trim.

    Free

    Security scanned

    About this skill

    Answer one question with numbers: what does this pack cost in context before I type a word?

    • Startup: loaded every session. CLAUDE.md / AGENTS.md, everything under rules/, and the name + description of every skill, agent and command.
    • On-demand: loaded only when a skill, agent or command triggers (its body, plus files shipped next to a skill).

    Startup is the number that matters. It is paid on every session whether or not the pack is used.

    1. Locate the pack folder (a cloned repo or ~/.claude). Ask the user if unclear.
    2. Run it (zero dependencies, Node 20+, nothing to install):
      npx rules-bench path/to/pack [path/to/other-pack ...]
      
      If npm is unavailable: git clone https://github.com/ferhatatagun/rules-bench and run node bin/rules-bench.mjs <pack>. Useful flags:
      • --skip-root when the pack's own CLAUDE.md describes the repo, not the pack.
      • --rules common,typescript to count only the rules/<subdir>s the user will actually install.
      • --window <n> for a non-default window (default 200000).
      • --exact for an exact startup count (needs ANTHROPIC_API_KEY).
      • --json for machine-readable output.
    3. Report a table: pack, startup tokens, % of window, on-demand total, average cost per triggered skill.

    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 3 days ago

    • Free with an Agensi account

    Requires

    Npmjs
    Node 20+
    rules-bench
    Node.js
    Git

    Permissions

    Allowed Hosts

    registry.npmjs.org

    Runs npx rules-bench on a local folder; read-only.

    Listed3 days ago
    Updated2 days ago

    What's inside

    Frequently Asked Questions

    Measures what a CLAUDE.md, rules folder or skill/agent pack costs in context before your first prompt (startup) versus only when triggered (on-demand), then tells you what to trim.

    This skill uses the open SKILL.md standard. Use it through an MCP-compatible agent connected to Agensi, and check the compatibility information on this page for confirmed agents and required capabilities.

    Connect your AI agent to Agensi through MCP, then ask it to use Context Cost Audit — what does your Claude Code pack cost? on your task. Your agent loads the latest approved version each time. Paid skills require a purchase or active subscription. Skill files are not available for download; your licence does not allow you to save, copy or export any part of a skill.

    New in Agents & Orchestration

    Free