More screenshots

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    Veroya AI Team Kit (for Claude Code)

    1

    A working, tested starter system for running a small team of Claude Code agents: agent-file generator, rule templates, critic gate and a daily report image script.

    $19

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Veroya AI Team Kit (for Claude Code)

    Veroya AI Team Kit (for Claude Code)

    Example session with this skill installed

    Set up a team called "WebFlow" with a mission to build a React landing page. I need a Lead Developer and a UI Critic role.

    • Read your context and instructions
    • Compiled the veroya ai team

    I have generated the configuration for the WebFlow team and prepared the agent files.

    Agent Team Scaffolding

    The following files have been generated in generated-agents/ and are ready to be moved to your .claude/agents/ directory:

    RoleFileModelResponsibility
    Lead Developerlead-developer.mdclaude-3-5-sonnetComponent architecture and React implementation
    UI Criticui-critic.mdclaude-3-5-sonnetReviewing accessibility, Tailwind consistency, and UX

    I have also prepared a CLAUDE.md update. Preview of the Team Rules section:

    
    ### WebFlow Team Rules
    - **Chain of Command**: Lead Developer implements; UI Critic must approve all frontend changes.
    - **Critic Gate**: No PR is considered complete without a "PASS" from the UI Critic role.
    - **Handoff Format**: Agents must append a # Handoff section to the project memory after every session.
    

    Next steps

    1. Run mv generated-agents/*.md .claude/agents/ to activate the roles.
    2. Review the suggested rules in CLAUDE.md before I finalize the write.
    3. Test the Lead Developer agent by asking it to initialize the Vite project.

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

    About this skill

    A working, tested starter system for running a small team of Claude Code agents: a generator that writes the 5 agent files, a report-image maker and a critic gate (a reviewer role every finished item goes through), all running on your own machine.

    What it is

    A starter kit for setting up a small team of Claude Code agents in your own project. It includes a script that turns a simple team file into ready-to-use agent files, rule templates for chain of command, safety, a critic gate and a handoff report, and a script that draws a daily report image from a JSON file you fill in.

    This kit is based on the setup we use ourselves, simplified into five generic example roles and written as templates. Every role, name and number in the kit is an example you replace with your own.

    What the skill does

    Connect Claude Code to Agensi, then ask for it ("set up my agent team", "generate agent files from team.json", "add a critic gate", "make the daily report image"). Your agent asks for your team name, mission and roles, writes your team file, runs the generator, writes a team-rules section for your CLAUDE.md from the kit's rule templates, and asks you before anything goes into .claude/agents/ or CLAUDE.md. The kit's files are not downloaded: Agensi serves them to your agent while the skill runs.

    What you get

    • Generate agent files from one team file: edit team.json, run one command, copy the agent files you want into .claude/agents/.
    • Rule templates your agent reads and applies: chain of command, safety, a critic gate that sends finished work to a reviewer role, a handoff report format, and usage-discipline rules to keep runs short. Your agent writes a version fitted to your team into your CLAUDE.md.
    • A daily report image from a JSON file with a small Python script, so you can see one summary at a glance.

    What's inside (14 files plus SKILL.md)

    • SKILL.md (the instructions your agent follows)
    • README.md, LICENSE.txt, THIRD-PARTY-NOTICES.md
    • team-generator/ make_team.py, team.json (5 example roles)
    • report-image/ make_report.py, sample_input.json
    • rules/ team-rules.md, critic-gate.md, project-memory-template.md
    • video-tools/ watch_local.py (wrapper only), INSTALL.md
    • voice-tools/ test_kokoro.py, INSTALL.md

    The video and voice folders hold a wrapper script, a test script and install notes only. The tools themselves are installed from their own upstream projects. The generator and the report script make no network calls. The video wrapper runs a separate tool you install yourself, and the voice test downloads its model files on first run; SKILL.md tells your agent to run neither unless you ask.

    Requirements

    • An Agensi account, with Claude Code connected to Agensi (the skill is delivered through that connection).
    • Claude Code, installed and signed in. It is a separate product with its own subscription or usage cost. This kit does not include it.
    • Python 3.9 or newer.
    • Pillow 10.1 or newer for the report image (pip install pillow).

    What this is not

    • Not a guarantee of any result. We make no claim that using it will earn money or save a set amount of time or cost.
    • Not business, legal or financial advice.
    • Not a hands-off system. You review everything. The Critic role is an AI reviewer and can be wrong.
    • Not free to run. Agents use your Claude usage.
    • Not bundling third-party tools. The video and voice tools are separate projects with their own licences.

    Known limitations

    • Tested as a locally installed Claude Code skill on Windows; not yet tested through the Agensi connection.
    • Model names in team.json (sonnet, haiku, opus) are aliases that Claude Code may change; check its current documentation.
    • The report script draws only what you put in the JSON file. It does not collect data for you.

    AI disclosure

    The scripts and text files in this kit were written with generative AI (Claude) and tested by running the scripts from a fresh unzip. The kit contains no AI-generated images or audio. The images on this page were drawn by Python scripts from text and from the kit's own output; no image model was used.

    Licence

    Through Agensi you get Agensi's buyer licence: permanent, personal and non-transferable, to use the skill through your agent (Agensi terms 8.2). What the skill makes for you (your agent files, your CLAUDE.md section, your report image) is yours to use. Provided as is, with no warranty of any kind.

    Not affiliated with or endorsed by Anthropic. Claude and Claude Code are trademarks of Anthropic, PBC.

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.

    Requires Python 3.9+. Image generation requires Pillow 10.1+. Critic roles require human review. Compatibility depends on Claude Code agent folder structures.

    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 buying a skill.

    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 days ago

    Accounts
    Agensi account
    Software
    Claude CodePython 3.9 or newer
    Optional
    Pillow 10.1 or newer

    Permissions

    Terminal / Shell
    Read Files
    Write Files
    Network Access
    Environment Variables
    Listed5 days ago

    What's inside

    Frequently Asked Questions

    This kit automates the scaffolding of multi-agent hierarchies in Claude Code, replacing manual prompt engineering with a standardized JSON schema for role definitions, critic gates, and reporting lines.

    The kit requires Python 3.9+ and is specifically designed for the Claude Code CLI environment, leveraging the `.claude/agents/` directory structure for agent persistence.

    Yes, the skill includes a Python-based visualization tool that uses the Pillow library to transform your team's JSON task logs into graphical daily progress reports for stakeholders.

    The kit generates a structured `CLAUDE.md` file that defines strict chain-of-command protocols and "Critic Gates," ensuring agents must pass a review phase before finalizing code changes.

    You receive the core team generation scripts, a library of role-definition templates, the Python visual reporting module, and the standardized project memory schema for agent handoffs.

    New in Agents & Orchestration