agentplat▸control runs deployments agents metering + agent
← agents

slack-support

k8s-job edit yaml logs manual run →
deployment slack-support-k8s · v1 · owner cloud-engineering · tenant cloud-engineering · model claude-sonnet-5

instructions

You are a Mozilla Cloud Engineering Slack support agent.
The trigger carries the mention text and the thread it came from (sourceContext.channel,
sourceContext.thread). Use the slack.thread.read tool to pull the full thread, then produce a
concise summary, findings with evidence, and a single proposed slack.message.post action that
drafts a helpful reply. Do NOT post anything yourself; only propose it. A human approves before
anything is sent to Slack.

how it is invoked (triggers)

slack-app-mention

A connector converts each source event into this agent's input. Manual and CLI runs are additional triggers.

capabilities

slack.thread.read slack.message.post

input schema

{
  "type": "object",
  "required": [
    "request"
  ],
  "properties": {
    "request": {
      "type": "string"
    },
    "sourceContext": {
      "type": "object"
    }
  }
}

output schema

{
  "type": "object",
  "required": [
    "summary",
    "findings"
  ],
  "properties": {
    "summary": {
      "type": "string"
    },
    "findings": {
      "type": "array"
    }
  }
}

limits

{
  "timeout_seconds": 900,
  "max_tool_calls": 25,
  "max_cost_usd": 5.0
}

capability bindings (deployment)

capabilitytransport / credential
slack.thread.read{"transport": "stub"}
slack.message.post{"credential_ref": "SLACK_BOT_TOKEN"}

approvals (deployment)

The real approval control: which action types require a human before the executor performs them.

actionpolicy
slack.message.postrequired