Envelope Team

Envelope Team

Design and generate .envelope.json AI agent team definition files — the open standard for building and deploying multi-agent teams on Envelope.

When to Use This Skill

• User asks to build, design, or generate an Envelope team
• User describes a workflow that could be automated by a team of AI agents
• User wants to create a .envelope.json file for deployment
• User needs help with agent hierarchy, access policies, or gates

What This Skill Does

1. Takes a natural-language description of the team's purpose 2. Designs the agent hierarchy (supervisor → managers → workers) 3. Generates a schema-valid .envelope.json file 4. Includes access policies, variables, secrets, and gates as appropriate 5. Explains any schema concepts the user asks about

How to Use

Basic Usage

Build me an Envelope team for customer support triage — one supervisor,
two agents that read Zendesk tickets, and a Slack notifier for escalations.

Create an Envelope team definition for outbound sales — an SDR manager
overseeing two SDRs who work HubSpot leads and a RevOps analyst.

With human-in-the-loop gates

Build an Envelope team that drafts marketing emails and pauses for
human approval before sending.

Example

User: "Build me an Envelope team for content moderation — a supervisor and two reviewer agents that check posts against community guidelines."

Output:

{
  "$schema": "https://schema.openenvelope.org/team/v1.json",
  "name": "Content Moderation Team",
  "slug": "content-moderation",
  "version": "1.0.0",
  "description": "A supervisor and two reviewer agents that check posts against community guidelines.",
  "visibility": "team",
  "category": "ops",
  "requiredVariables": ["companyName"],
  "requiredSecrets": ["MODERATION_API_KEY"],
  "metadata": { "generatedBy": "Claude Code · openenvelope.org" },
  "agents": [
    {
      "key": "supervisor",
      "name": "Moderation Supervisor",
      "role": "supervisor",
      "capabilities": ["Route content to reviewer agents", "Aggregate decisions", "Escalate edge cases"],
      "model": "anthropic:claude-sonnet-4-5",
      "systemPrompt": "You supervise the content moderation team at {{companyName}}. Delegate each item to a reviewer and consolidate their verdicts."
    },
    {
      "key": "policy-reviewer",
      "name": "Policy Reviewer",
      "role": "reviewer",
      "capabilities": ["Check content against community guidelines", "Flag policy violations"],
      "model": "anthropic:claude-haiku-3-5",
      "systemPrompt": "You review content for policy violations at {{companyName}}. Return a verdict of approve, flag, or remove with a reason.",
      "reportsToKey": "supervisor",
      "accessPolicy": {
        "accessPolicyVersion": "1",
        "rules": [
          { "host": "api.moderation-service.com", "action": "allow" },
          { "host": "*", "action": "deny" }
        ]
      }
    },
    {
      "key": "spam-reviewer",
      "name": "Spam Reviewer",
      "role": "reviewer",
      "capabilities": ["Detect spam, bot activity, and duplicate content"],
      "model": "anthropic:claude-haiku-3-5",
      "systemPrompt": "You detect spam and bot activity at {{companyName}}. Return a verdict of approve, flag, or remove with a confidence score.",
      "reportsToKey": "supervisor"
    }
  ]
}

Schema Reference

Top-level fields

Agent fields

Hierarchy rule

Exactly one agent should have no reportsToKey — this is the supervisor that receives the initial task. All others reference their manager's key.

Access policy

"accessPolicy": {
  "accessPolicyVersion": "1",
  "rules": [
    { "host": "api.example.com", "action": "allow" },
    { "host": "*", "action": "deny" }
  ]
}

Human-in-the-loop gates

"gates": [
  {
    "name": "approval-check",
    "type": "approval",
    "trigger": { "afterAgent": "drafter" },
    "onApprove": "continue",
    "onReject": "halt"
  }
]

Tips

• Use anthropic:claude-sonnet-4-5 for supervisors and managers; anthropic:claude-haiku-3-5 for leaf agents doing repetitive work
• Put anything org-specific (company name, support email, Slack channel) in requiredVariables
• End access policy rules with { "host": "*", "action": "deny" } for a strict allowlist
• Add gates before any agent that sends emails, posts publicly, or makes purchases
• Validate before deploying: npx @openenvelope/schema validate ./team.envelope.json

Links

• openenvelope.org — deploy and run teams
• schema.openenvelope.org/team/v1.json — JSON schema
• npmjs.com/package/@openenvelope/schema — TypeScript types + validator
• openenvelope.org/mcp — run teams via MCP from Claude.ai or ChatGPT