OpenAI Agents SDK Production Specialist Agent

## Content

OpenAI Agents SDK Production Specialist Agent is a reusable agent prompt for
designing and reviewing production applications built with the OpenAI Agents
SDK. It focuses on the parts that matter after a prototype starts handling real
users: agent loop ownership, tool contracts, handoffs, guardrails, sessions,
tracing, MCP integrations, sandbox execution, deployment limits, and privacy
controls.

Use this agent when a team is deciding whether to use the Agents SDK or the
Responses API directly, preparing an SDK-based service for production, reviewing
a pull request that adds tool or handoff behavior, or investigating an incident
where traces, session state, or tool calls need to be reconstructed.

## Features

- Source-backed workflow for OpenAI Agents SDK architecture and release review.
- Coverage of SDK primitives: `Agent`, `Runner`, tools, handoffs, guardrails,
  sessions, tracing, MCP servers, sandbox agents, realtime agents, and voice
  agents when relevant.
- Decision framework for Agents SDK versus direct Responses API usage.
- Tool and MCP review checklist for least privilege, validation, idempotency,
  timeouts, side effects, and auditability.
- Handoff review checklist for input filters, transcript minimization, nested
  handoff behavior, and final-output ownership.
- Session review checklist for persistence, tenant isolation, retention,
  storage backend choice, and history limits.
- Tracing review checklist for sensitive-data capture, trace processors, export
  timing, group IDs, long-running workers, and zero-data-retention constraints.
- Release output contract with blockers, non-blocking improvements, validation
  plan, and final release decision.

## Use Cases

- Review a production OpenAI Agents SDK pull request before it can call
  write-capable tools or MCP servers.
- Design a multi-agent workflow with clear handoff ownership and filtered
  context.
- Add input, output, and tool guardrails without assuming they apply to every
  step of the workflow.
- Choose a session backend and retention model for a multi-turn agent.
- Debug a production issue by mapping traces, spans, tool calls, and handoffs
  back to the SDK runtime.
- Prepare a deployment checklist for background workers, trace flushing,
  cost limits, rollback, and canary monitoring.

## Source Notes

- OpenAI's Agents SDK docs describe the SDK as a higher-level runtime around
  OpenAI model calls with agents, tools, handoffs, guardrails, sessions,
  tracing, MCP server tool calling, sandbox agents, realtime agents, and voice
  agents.
- The Python docs document built-in tracing, including generation spans,
  function spans, guardrail spans, handoff spans, sensitive-data controls, and
  trace disabling options.
- The handoffs docs describe input filters, nested handoff history behavior,
  and the scope of input/output guardrails.
- The sessions docs describe built-in memory, session backends, persistence
  behavior, and cases where sessions should not be combined with server-managed
  continuation IDs.
- The official `openai/openai-agents-python` repository provides the source
  package, examples, docs, and project metadata for the Python SDK.

## Duplicate Check

Before drafting this entry, the current upstream content tree and live open PRs
were checked for OpenAI Agents SDK, `openai-agents-python`, `openai-agents`,
`openai.github.io/openai-agents-python`, the OpenAI Agents SDK platform guide,
and source-specific title variants. Existing AgentOps and TruLens entries only
mention OpenAI Agents SDK as an integration; no dedicated agents entry or open
PR covers this production specialist agent.

## Editorial Disclosure

Submitted as a community agent entry by `oktofeesh1`. This listing is based on
OpenAI's official Agents SDK documentation and repository, with no paid
placement or affiliate relationship.