Dual Session Exec

    1

    Dual-session plan/execute skill: main session coordinates and independently reviews; a foreground executor implements and self-tests an approved plan.

    Free

    Security scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Dual Session Exec

    Example session with this skill installed

    Use dual-session-exec to plan and implement a small feature: add input validation to my signup form. Plan first, then hand the approved plan to an executor session.

    • Read your context and instructions
    • Compiled the dual session exec

    Summary

    The main session produced a plan, you approved it, and a foreground executor implemented and self-tested it. The main session then independently reviewed the result.

    StepOwnerStatus
    Plan: validation rules for email, password, nameMain sessionApproved
    Implement validators and error messagesExecutorDone
    Self-test: unit tests for each ruleExecutor12/12 passing
    Independent review of diff and testsMain sessionNo blocking issues
    export const isValidEmail = (v: string) => /^[^@\s]+@[^@\s]+\.[^@\s]+$/.test(v);
    

    Next steps

    • Merge the change after reviewing the diff
    • Run the full test suite in CI

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

    About this skill

    Dual-session plan/execute skill: the main session coordinates and independently reviews the work, while a foreground executor session implements and self-tests an approved plan.

    Source: https://github.com/ChangyuZeng/dual-session-exec

    Requires Python 3.11+ for scripts. State locks and job daemons require POSIX. Communication depends on host-provided tools or manual user relay. No automatic session archiving.

    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 12 days ago

    • Free with an Agensi account

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Listed12 days ago

    What's inside

    Frequently Asked Questions

    This skill solves 'context drift' by splitting complex tasks into two sessions: a primary coordinator that maintains the architecture and a worker that executes code. By using frozen contracts and state files, it prevents the AI from losing track of requirements during long development cycles.

    The skill is optimized for POSIX environments and requires Python 3.11+ for its state-tracking scripts. It is designed to work with Codex desktop tools, WorkBuddy, and any agent interface that supports sub-agent messaging or manual user-relay.

    You receive the complete orchestration logic, the `contract.md` and `state.json` templates, and the Python utility scripts (`session_config.py`, `review_receipt.py`) used to synchronize progress between sessions.

    Yes, while it excels at automated handovers in supported environments, it includes a 'manual user-relay' workflow. This allows you to act as the bridge between two separate chat instances while the skill manages the state and verification logic.

    The primary session uses the `contract.md` file as a source of truth to perform independent validation. It checks the worker's output against pre-defined fingerprints and exit criteria before marking a task as complete.

    New in Agents & Orchestration