> ## Documentation Index
> Fetch the complete documentation index at: https://apie.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Production release gate

> Walkthrough — pipeline session with LLM risk summary and high-risk CI/CD MCP trigger.

Before deploying to production, your release gate agent collects evidence, summarizes risk via an LLM tool call, and may trigger a CI/CD pipeline. This recipe models that as a pipeline session with explicit high-risk metadata.

**Source:** [production-release-gate-loop.ts](https://github.com/apie-sh/javascript-sdk/blob/main/examples/production-release-gate-loop.ts) · [production\_release\_gate\_loop.py](https://github.com/apie-sh/python-sdk/blob/main/examples/production_release_gate_loop.py)

## Scenario

A release gate orchestrator validates rollout readiness. It calls an LLM tool to summarize risk, then triggers a CI/CD pipeline via MCP. Both actions need production metadata for guardrails.

## Step 1 — Pipeline session

<CodeGroup>
  ```ts TypeScript theme={null}
  await apie.withSession(
    {
      kind: "pipeline",
      inputSummary: "Validate production rollout readiness",
      metadata: { workflow: "release_gate" },
    },
    async (session) => { /* orchestrator run inside */ },
  );
  ```

  ```python Python theme={null}
  apie.with_session(
      {
          "kind": "pipeline",
          "inputSummary": "Validate production rollout readiness",
          "metadata": {"workflow": "release_gate"},
      },
      lambda session: None,  # orchestrator run inside
  )
  ```
</CodeGroup>

## Step 2 — Orchestrator run with step metadata

<CodeGroup>
  ```ts TypeScript theme={null}
  await apie.withRun(
    {
      sessionId: session.id,
      stepKey: "orchestrator",
      stepName: "Release gate orchestrator",
      stepIndex: 0,
      inputSummary: "Collect rollout evidence and risky actions",
    },
    async (run) => { /* tool calls inside */ },
  );
  ```
</CodeGroup>

## Step 3 — LLM tool call (medium risk)

<CodeGroup>
  ```ts TypeScript theme={null}
  await withOpenAIToolCall(apie, {
    runId: run.id,
    toolName: "summarize_release_risk",
    arguments: { service: "api", environment: "production" },
    resourceType: "work_item",
    riskLevel: "medium",
  }, async () => ({ summary: "No blocker found in latest incident feed." }));
  ```

  ```python Python theme={null}
  from apie import with_openai_tool_call

  with_openai_tool_call(apie, {
      "runId": run.id,
      "toolName": "summarize_release_risk",
      "arguments": {"service": "api", "environment": "production"},
      "resourceType": "work_item",
      "riskLevel": "medium",
  }, lambda: {"summary": "No blocker found."})
  ```
</CodeGroup>

## Step 4 — MCP CI/CD trigger (high risk)

<CodeGroup>
  ```ts TypeScript theme={null}
  await withMcpToolCall(apie, {
    runId: run.id,
    sessionId: session.id,
    server: "internal-cicd",
    tool: "trigger_pipeline",
    actionType: "execute",
    resourceType: "pipeline_run",
    environment: "production",
    riskLevel: "high",
    resourceTarget: "payments-service",
  }, async () => ({ accepted: true, runId: "pipe_123" }));
  ```

  ```python Python theme={null}
  from apie import with_mcp_tool_call

  with_mcp_tool_call(apie, {
      "runId": run.id,
      "sessionId": session.id,
      "server": "internal-cicd",
      "tool": "trigger_pipeline",
      "actionType": "execute",
      "resourceType": "pipeline_run",
      "environment": "production",
      "riskLevel": "high",
      "resourceTarget": "payments-service",
  }, lambda: {"accepted": True, "runId": "pipe_123"})
  ```
</CodeGroup>

### What you'll see

A pipeline session replay with orchestrator run, LLM tool call (medium risk), and MCP pipeline trigger (high risk). In Monitor mode, guard evaluations show what would be blocked in Enforce mode.

## Escalate to enforcement

1. [Declare capabilities](/boundaries/declare-capabilities) for `trigger_pipeline` and `summarize_release_risk`
2. [Enable guardrail templates](/guardrails/enable-guardrail-templates) for prod-deploys
3. Switch to [Enforce mode](/guardrails/enforce-guardrails)

## Next steps

<CardGroup cols={2}>
  <Card title="Multi-agent handoff" icon="diagram-project" href="/recipes/multi-agent-handoff">
    Orchestrator delegates to worker.
  </Card>

  <Card title="MCP instrumented client" icon="code" href="/mcp/instrumented-client">
    In-process MCP setup.
  </Card>
</CardGroup>