Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Zênite - Infinity Canvas - Agent Team: split, delegate, merg

    1

    A method for running several AI coding agents on one project without them stepping on each other.

    $4.99

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Zênite - Infinity Canvas - Agent Team: split, delegate, merg

    Zênite - Infinity Canvas - Agent Team: split, delegate, merg

    Example session with this skill installed

    Use the Agent Team skill. I have 3 agents available (Claude Code, Codex and Cursor).

    Task: add CPF validation to the signup form, log every failed signup attempt to a new signup_attempts table, and write the migration. Repo is a Node/Express API with Prisma. Tests run with npm test.

    Split the work, write the brief for each agent and tell me how to merge it at the end.

    • Read your context and instructions
    • Compiled the z nite

    Summary

    Three pieces that touch different files, one agent each, merged through Git at the end. Coordinator: you. Estimated wall time: about 25 minutes in parallel instead of 70 in sequence.

    Split

    PieceOwnerMay touchMust not touchDone when
    A. CPF validationClaude Codesrc/validators/cpf.ts (new), src/routes/signup.tsprisma/, src/services/log*npm test passes; 11 digits, check digits and repeated-digit CPFs rejected
    B. Migration + modelCodexprisma/schema.prisma, prisma/migrations/*src/routes/, src/validators/npx prisma migrate dev runs clean; model SignupAttempt with email, reason, ip, createdAt
    C. Attempt loggingCursorsrc/services/signupLog.ts (new), src/routes/signup.ts (one call only)prisma/schema.prismafailed attempt writes one row; unit test with a mocked Prisma client

    Pieces A and C both touch src/routes/signup.ts. To avoid a conflict, C adds a single line await logAttempt(req, reason) inside the existing catch, nothing else. A owns the rest of the file.

    Brief for each agent

    Agent: Claude Code
    Branch: feat/cpf-validation
    Goal: reject invalid CPF on POST /signup with a 400 and the message "CPF inválido".
    May touch: src/validators/cpf.ts (create), src/routes/signup.ts.
    Must not touch: prisma/, src/services/.
    Done when: npm test passes and these cases are covered: 11 digits only, valid check digits, all-same-digit CPFs rejected.
    When done: commit as "Validate CPF on signup" with a line starting "Why:", then reply with the test output.
    
    Agent: Codex
    Branch: feat/signup-attempts-table
    Goal: add a SignupAttempt model and its migration.
    May touch: prisma/schema.prisma, prisma/migrations/.
    Must not touch: src/.
    Done when: npx prisma migrate dev runs clean and the generated client has SignupAttempt.
    When done: commit as "Add signup_attempts table" with a "Why:" line and paste the migration SQL.
    
    Agent: Cursor
    Branch: feat/signup-attempt-log
    Goal: log every failed signup to SignupAttempt.
    May touch: src/services/signupLog.ts (create), src/routes/signup.ts (add one call in the existing catch block only).
    Must not touch: prisma/schema.prisma.
    Done when: a failing signup writes one row; test uses a mocked Prisma client.
    When done: commit as "Log failed signup attempts" with a "Why:" line and reply with the diff of signup.ts.
    

    Merge order and check

    1. Merge B first (the table has to exist before C's code runs).
    2. Merge A.
    3. Merge C last. Expect one small conflict in src/routes/signup.ts: keep A's validation block and C's single logAttempt line.
    4. Run npx prisma migrate dev, then npm test.
    5. Read the full diff once before pushing. Reject any change outside the "may touch" list.

    Next steps

    • Send each brief to its agent as the first message, unchanged.
    • Do not start step 3 until each agent has replied with its proof (test output, SQL, diff).
    • If an agent reports "done" without proof, send it back with: "Show me the test output."

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

    About this skill

    Your agent loads this skill before a task that is too big for one context window or has parts that can run side by side. It then works the way a good tech lead does:

    Split the request into pieces that touch different files. One piece, one owner, one verifiable result. Write a brief for each agent: goal, files it may touch, files it must not touch, and how to prove it is done. Assign roles: a coordinator that only distributes and checks, executors that do the work, and a reviewer that did not write the code. Use Git as shared memory: small commits that say why, one branch per agent, merge at the end. Merge and verify with a checklist: run the tests, read the diff, find conflicts before accepting.

    It also names the mistakes to avoid: two agents editing the same file, a vague brief (“improve this”), and accepting “done” without proof.

    What you get

    The briefing template, ready to paste. The integration checklist. The commit-message format for agents (title, why, what is pending).

    Works with Claude Code, Codex, Cursor, Gemini CLI and any agent that reads SKILL.md.

    Not for small, single-file tasks. One agent is faster there.

    Built by the author of Zênite, a Windows canvas where terminal agents work together with a coordinator. This skill is the method Zênite runs on its own.

    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 1 day ago

    Requires

    Git

    Permissions

    Network Access
    Listed1 day ago

    What's inside

    Frequently Asked Questions

    A method for running several AI coding agents on one project without them stepping on each other.

    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 Zênite - Infinity Canvas - Agent Team: split, delegate, merg 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