Agent Project Manager

    1

    Route each task to the least expensive model that can handle it, with a clear score and focused handoff.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    About this skill

    Not every task needs a frontier model. Agent Project Manager scores work by difficulty, chooses from local, low-cost, mid-tier, and frontier models, passes along only the skills a subtask needs, and gives the next agent a finite prompt. You control the tiers and prices in a plain YAML file, so the routing reflects the models and backends you actually use.

    Requires manual price updates in tiers.yaml. User must have local servers (Ollama/LM Studio) running for local tier routing. Subagent execution depends on the host environment's capabilities.

    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 9 days ago

    Requires

    ailoadout
    Node.js

    Permissions

    Terminal / Shell
    Network Access
    Write Files
    Listed9 days ago

    What's inside

    Frequently Asked Questions

    It uses a custom difficulty rubric to analyze subtasks, automatically routing simple tasks to local or low-cost models while reserving expensive frontier models for complex architectural work.

    The skill includes pre-configured YAML and Markdown templates that you can customize to match your preferred local (Ollama/LM Studio) or cloud-based API providers.

    If a task fails its quality check, the agent automatically re-evaluates the difficulty and escalates the subtask to a higher-tier model for verification or a complete rewrite.

    You receive the core logic for task scoring and routing, a suite of configuration templates (rubric, tiers, and prompts), and a reporting tool that generates a cost-savings receipt for every project.

    Yes, you can define specific data types to be handled exclusively by local backends in your configuration, ensuring sensitive code never leaves your local environment.

    New in Agents & Orchestration