---
title: "Guild vs OpenAI Agents API"
description: "Guild vs the OpenAI Agents API: sessions, sandboxes, tools, credentials, schedules, models, and pricing compared, plus how to move an agent to Guild."
canonical: "https://www.guild.ai/compare/openai-agents-api"
---

# Guild vs OpenAI Agents API

Guild vs the OpenAI Agents API: sessions, sandboxes, tools, credentials, schedules, models, and pricing compared, plus how to move an agent to Guild.

The OpenAI Agents API runs OpenAI's Codex agent harness for you, with durable sessions and an optional sandbox. Guild runs agents on models from several providers, OpenAI included, keeps credentials behind a proxy, and starts agents from Slack, GitHub, a schedule, or an API call.

Last verified 2026-11-06. Across 35 features, Guild leads on 16, OpenAI Agents API on 7, and 12 are level. Every cell below names its source.

## When to choose OpenAI Agents API

- You only use OpenAI models and want OpenAI to run the agent for you.
- Your agents do long coding or file work and need a sandbox, hosted or your own.
- You want to steer an agent while it runs, or use computer use.
- You want no platform fee: model tokens and tools at API rates.

## When to choose Guild

- You want agents on models from several providers, with fallbacks.
- You want managed integrations, or new ones from your own OpenAPI spec.
- You want agents to start from Slack or GitHub events, not only the API and a cron.
- You want to share agents and skills publicly on an Agent Hub others can install or fork.

## What OpenAI Agents API and Guild are

**OpenAI Agents API.** The Agents API is Codex as a managed service. A session is a durable instance OpenAI runs and recovers, with tools, MCP servers, skills, sub-agents, and an optional sandbox hosted by OpenAI or by you. Credentials sit in vaults, and automations run sessions on a cron. It's in public beta, runs OpenAI models only, and is US data residency only.

**Guild.** Guild runs agents built several ways, on models from several providers. Integrations come from a managed catalog or your own OpenAPI spec, brokered by a credential proxy with ALLOW/DENY rules per operation, and agents start from schedules, Slack, GitHub, or API calls. Teams share agents on the Agent Hub.

|   | OpenAI Agents API | Guild |
| --- | --- | --- |
| Category | Hosted agent runtime API | One platform to run and control AI agents |
| License | Proprietary service; Codex CLI and harness Apache-2.0 | Proprietary platform; CLI published on npm as @guildai/cli |
| Pricing | No Agents API fee; model tokens and tools at API rates; hosted sandboxes $0.03–$0.48 per 20-minute session | No public price list; plans priced with the Guild team; model usage metered at list price |
| Maturity | Public beta (2026-09-10) | Generally available |
| Docs | https://platform.openai.com/docs | https://docs.guild.ai |

## Concept map: OpenAI Agents API to Guild

| Concept | OpenAI Agents API | Guild | Note |
| --- | --- | --- | --- |
| The agent | Saved agent (POST /v1/agents: model, instructions, tools) | A Guild agent: Native (PROMPT.md + guild.yaml) for instructions plus tools, TypeScript for code |   |
| Sandbox with shell and files | Environment openai_hosted (Bash, apply-patch, /workspace) | An OpenClaw agent (AGENTS.md + skills/) in Guild's container, with shell, git, and file editing |   |
| A run | Session and turns (/v1/agents/sessions) | A Guild session and its event log (guild session) |   |
| Custom tools | function tools your app executes | Tools inside a TypeScript agent, or a custom integration from an OpenAPI spec (Porter) |   |
| MCP servers | mcp tools with allowed_tools | A custom Guild integration from the MCP server, with credential policies per operation |   |
| Credentials | Vaults (static_bearer, mcp_oauth, environment_variable) | Credentials behind the credential proxy (account, workspace, member, or agent scope) | Both keep secrets out of the agent's reach. |
| Typed output | agent.text.format.schema | outputSchema (zod) on a TypeScript agent, plus a typed inputSchema |   |
| Sub-agents | multi_agent subagents | Sub-agents with guildAgentTool |   |
| Skills and plugins | Skills and plugins (capability_directories, .codex-plugin) | Guild skills on the Hub (guild skill create), public or private |   |
| Schedules | Automations (/v1/agents/automations, cron + timezone) | guild trigger create --type time |   |
| Starting from events | POST /v1/agents/sessions and signed webhooks | guild workspace chat --once, API triggers, and integration-event triggers |   |
| Approvals | review: "auto_review" and computer-use approvals | Approval gates set in the Guild UI, per agent and per action |   |
| Tracing | Tracing (Logs > Agents, OTLP export) | Sessions, Insights, and Spend Explorer |   |
| Spend limits | Org and project spend limits | Spend budgets and token limits |   |

## Feature by feature: OpenAI Agents API vs. Guild

### Authoring

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Code-first TypeScript agents | **Partial.** The official `openai` npm SDK exposes `client.beta.agents` (sessions, agents, environments, automations). An agent is declarative config (model, instructions, tools), and your TypeScript app handles function tools. OpenAI does not host code-first TS agent logic. `@openai/codex-sdk` drives local Codex threads only. ([source](https://developers.openai.com/api/docs/guides/agents-api/overview), checked 2026-11-06) | **Yes.** @guildai/agents-sdk: agent(), llmAgent(), and the task API (task.llm, task.tools, task.ui, task.gather). ([source](https://docs.guild.ai/packages/agents-sdk.md), checked 2026-10-23) |
| Python / LangGraph agents | **Partial.** Python is supported through `openai` (`client.beta.agents.sessions.create`) and `pip install openai-codex` (Codex SDK). There is no LangGraph integration; the agent loop is OpenAI's Codex harness. ([source](https://developers.openai.com/api/docs/guides/agents-api/quickstart), checked 2026-11-06) | **Yes.** LangGraph agent type: langgraph.json + graph.py using guild_model() and guild_tools(). LangChain v1, Python 3.13, no extra pip packages. ([source](https://docs.guild.ai/guide/agent-types.md), checked 2026-10-23) |
| Instruction-only (no-code) agents | **Partial.** A saved agent is just model + instructions + tools (`POST /v1/agents`), so no code project is needed, but you still create and run it through API calls. Agent Builder (visual canvas) is deprecated and shuts down 2026-11-30. ChatGPT workspace agents are OpenAI's separate no-code path. ([source](https://developers.openai.com/api/docs/guides/agent-builder), checked 2026-11-06) | **Yes.** Native agents: PROMPT.md + guild.yaml, no container, ready immediately. ([source](https://docs.guild.ai/guide/agent-types.md), checked 2026-10-23) |
| Several agent formats on one platform | **No.** One runtime: the managed Codex harness. OpenAI's other agent runtimes (the Agents SDK and Responses API) run in your own application, not on the Agents API. ([source](https://developers.openai.com/api/docs/guides/agents), checked 2026-11-06) | **Yes.** Native, TypeScript, Goose, OpenClaw, and LangGraph agent types, plus ADK (CLI 0.28). ([source](https://docs.guild.ai/guide/agent-types.md), checked 2026-10-23) |
| Typed agent input/output | **Partial.** Output can be constrained with a JSON schema (`agent.text.format.schema`), and function tools declare JSON-schema parameters. Session input is messages; there is no declared input schema. ([source](https://developers.openai.com/api/docs/guides/agents-api/sessions), checked 2026-11-06) | **Yes.** Zod inputSchema and outputSchema on agent(); root input must be z.object. ([source](https://docs.guild.ai/packages/agents-sdk.md), checked 2026-10-23) |
| Repo / workspace agents (shell, git, files) | **Yes.** With an `openai_hosted` or `self_hosted` environment the agent gets built-in Bash and apply-patch tools, a `/workspace` filesystem, and artifacts published from `/workspace/outputs`. Codex Cloud adds the repo/diff/PR workflow in ChatGPT. ([source](https://developers.openai.com/api/docs/guides/agents-api/architecture), checked 2026-11-06) | **Yes.** OpenClaw agents (AGENTS.md + skills/) run in the guildai~lobsterpot container with shell, git, and file editing. ([source](https://docs.guild.ai/guide/agent-types.md), checked 2026-10-23) |
| Sub-agents and agent-to-agent calls | **Yes.** `agent.multi_agent.enabled` gives the harness tools to create, message, wait for and interrupt subagents (`max_concurrent_subagents` defaults to 6). Subagents share the environment and inherit MCP tools but cannot use function tools. ([source](https://developers.openai.com/api/docs/guides/agents-api/multi-agent), checked 2026-11-06) | **Yes.** guildAgentTool calls owner~agent; Native, Goose, and OpenClaw agents declare sub_agents in guild.yaml. ([source](https://docs.guild.ai/guide/guild-yaml.md), checked 2026-10-23) |

### Tools & integrations

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Managed integrations catalog | **No.** There is no managed connector catalog for Agents API sessions. You connect your own HTTP/stdio MCP servers or upload plugin ZIPs. The built-in Responses connectors are labeled legacy. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/mcp), checked 2026-11-06) | **Yes.** 37 non-model integrations (GitHub, Slack, Gmail, Linear, Jira, Notion, Zendesk, …) plus 8 model publishers. ([source](https://docs.guild.ai/integrations/overview.md), checked 2026-10-23) |
| Use MCP servers as tools | **Yes.** HTTP MCP servers can connect from OpenAI (`connection_origin: "service"`, default) or from the session environment, and stdio servers run in the environment. Options include `allowed_tools`, `required`, and vault-backed auth. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/mcp), checked 2026-11-06) | **Yes.** Build custom integrations from MCP servers with guild integration. ([source](https://docs.guild.ai/integrations/overview.md), checked 2026-10-23) |
| Platform exposed over MCP | **No.** The Agents API has no documented MCP endpoint for exposing agents. Codex's own `codex mcp-server` command and binary were removed in favor of the JSON-RPC app server. ([source](https://learn.chatgpt.com/docs/mcp-server), checked 2026-11-06) | **Yes.** Guild is an MCP server: https://api.guild.ai/mcp (OAuth + PKCE) or local `guild mcp`. ([source](https://docs.guild.ai/api-reference/mcp.md), checked 2026-10-23) |
| Custom APIs from OpenAPI | **No.** No OpenAPI-to-tools import is documented. You hand-write function tools (your app executes them) or use MCP servers. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/functions), checked 2026-11-06) | **Yes.** guild integration builds tools from OpenAPI specs; Porter drafts integrations from public APIs. ([source](https://docs.guild.ai/integrations/overview.md), checked 2026-10-23) |
| Secrets never reach the agent | **Yes.** Vaults hold `static_bearer`/`mcp_oauth` credentials that OpenAI uses for MCP calls. `environment_variable` credentials give the hosted sandbox only a placeholder, which a proxy swaps for the real secret on allowed hosts. Self-hosted sandboxes need your own proxy. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/vaults), checked 2026-11-06) | **Yes.** A credential proxy injects auth server-side; GitHub uses short-lived GitHub App tokens. ([source](https://docs.guild.ai/platform/credentials.md), checked 2026-10-23) |
| Per-operation credential policies | **Partial.** Each environment credential has `allowed_hosts` (HTTPS 443/8443 only), and each MCP tool can set `allowed_tools`. There is no per-operation allow/deny policy attached to a credential. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/vaults), checked 2026-11-06) | **Yes.** ALLOW/DENY rules by operation, resource, agent, and workspace; DENY wins. ([source](https://docs.guild.ai/platform/credential-policies.md), checked 2026-10-23) |
| Act as the invoking user | **Partial.** Per-user OAuth grants can be stored as `mcp_oauth` vault credentials (with refresh) and attached per session through `vault_ids`, but your application must run the provider consent flow. ([source](https://developers.openai.com/api/docs/guides/agents-api/tools/vaults), checked 2026-11-06) | **Yes.** Member-scoped credentials plus identity mapping for Slack and GitHub users. ([source](https://docs.guild.ai/platform/credentials.md), checked 2026-10-23) |

### Runtime & triggers

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Managed hosted runtime | **Yes.** OpenAI runs the Codex harness (agent loop, sessions, compaction, recovery) and can provision a per-session Linux sandbox (`environment.type: "openai_hosted"`). ([source](https://developers.openai.com/api/docs/guides/agents-api/overview), checked 2026-11-06) | **Yes.** Guild hosts and runs all agent types. ([source](https://docs.guild.ai/platform/security-architecture.md), checked 2026-10-23) |
| Self-host option | **Partial.** Only execution can be self-hosted: you run the `codex exec-server` executor on your own compute (laptop, container, VPC, or providers such as E2B, Modal, Daytona, Cloudflare, Vercel, AWS Lambda MicroVMs). The harness and inference always run at OpenAI, or in Amazon Bedrock with the preview Bedrock Managed Agents. ([source](https://developers.openai.com/api/docs/guides/agents-api/environments/self-hosted), checked 2026-11-06) | **No.** Hosted only; no self-host option documented. ([source](https://docs.guild.ai/platform/security-architecture.md), checked 2026-10-23) |
| Sandboxed, network-controlled execution | **Yes.** Each session gets its own hosted sandbox (small 1 vCPU/1 GB, medium 2/4 GB, large 4/16 GB). Network can be `enabled`, `disabled` or `restricted` to 1-100 exact hosts. Hosted stdio MCP requires `enabled`. ([source](https://developers.openai.com/api/docs/guides/agents-api/environments/openai-hosted), checked 2026-11-06) | **Yes.** Containers lose network access before the agent runs; only Guild proxies can reach out. ([source](https://docs.guild.ai/platform/security-architecture.md), checked 2026-10-23) |
| Durable long-running runs | **Yes.** Sessions are durable and resumable: OpenAI manages orchestration, compaction and recovery, and a session can wait in `requires_action` for function results. Limits: a hosted sandbox is deleted after 1 hour without activity (not configurable), and a self-hosted reconnect wait times out after 5 minutes. ([source](https://developers.openai.com/api/docs/guides/agents-api/environments/openai-hosted), checked 2026-11-06) | **Partial.** Multi-turn sessions; they can be interrupted but not resumed. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Schedules (cron) | **Yes.** `/v1/agents/automations` supports `periodic` (5-field cron + IANA timezone) and `once` schedules that target new or existing sessions. An optional `automations` tool lets agents schedule their own follow-ups. ([source](https://developers.openai.com/api/docs/guides/agents-api/sessions/automations), checked 2026-11-06) | **Partial.** CLI supports --frequency CRON --cron-expression; docs list hourly–monthly presets. ([source](https://docs.guild.ai/platform/triggers.md), checked 2026-10-23) |
| Webhook, event & API triggers | **Yes.** A REST/SDK API starts sessions and sends input. Signed outbound webhooks fire for `agent.session.created`, `action_required`, `in_progress`, `idle` and `failed`. There are no native inbound third-party event triggers. ([source](https://developers.openai.com/api/docs/guides/agents-api/sessions/webhooks), checked 2026-11-06) | **Yes.** Event triggers (Slack mention, GitHub PR, Zendesk ticket, …) and API triggers. ([source](https://docs.guild.ai/platform/triggers.md), checked 2026-10-23) |
| Chat channels (Slack, Teams, …) | **Partial.** The Agents API has no native channels; OpenAI ships a cookbook Slack bot that you host. Separately, Codex Cloud is reachable from @ChatGPT in Slack/Teams (cloud delegation) and Codex in Linear. ([source](https://developers.openai.com/api/docs/guides/agents-api/overview), checked 2026-11-06) | **Yes.** Slack event triggers; Discord, Telegram, and Twilio integrations. ([source](https://docs.guild.ai/integrations/overview.md), checked 2026-10-23) |

### Control & governance

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Per-tool-call approval gate | **Partial.** Only computer-use origin-access and sign-in requests (`computer_use_approval_request`) pause for a human, plus function tools that your app gates itself. Other risky actions are either judged by an OpenAI reviewer model (`review: "auto_review"`) or just run (`none`). They are not sent to your app for manual approval. ([source](https://developers.openai.com/api/docs/guides/agents-api/sessions/approvals), checked 2026-11-06) | **Yes.** Approval gates are set in the Guild UI, per agent and per action: when the agent reaches a gated action, the run pauses until a designated human approves. No code needed. ([source](https://www.guild.ai/product/ai-governance), checked 2026-10-23) |
| Roles & access control | **Yes.** Platform organization and project roles, custom roles and SCIM-synced groups apply. API keys are scoped (`api.agents.read/write`, `api.vaults.read/write`), and environment keys can only connect executors. ([source](https://developers.openai.com/api/docs/guides/rbac), checked 2026-11-06) | **Partial.** Admin and Member org roles, restricted workspaces, and scoped API keys. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Audit log | **Partial.** The Admin API Audit Logs endpoint lists organization user actions and configuration changes. Coverage of Agents API resources isn't documented; session items and traces give a per-run history. ([source](https://developers.openai.com/api/docs/guides/admin-apis), checked 2026-11-06) | **Yes.** Immutable, CSV-exportable audit log (admins). ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Runtime guardrails & scanning | **Partial.** Safety comes from the auto-review reviewer model, network restrictions, credential brokering and platform safety checks. There is no configurable PII or content-guardrail layer in the Agents API; guardrails are an Agents SDK feature. ([source](https://developers.openai.com/api/docs/guides/agents-api/environments/security), checked 2026-11-06) | **No.** No prompt or tool-call scanning documented; isolation comes from the network-less runtime and credential policies. ([source](https://docs.guild.ai/platform/security-architecture.md), checked 2026-10-23) |
| Spend budgets & usage limits | **Yes.** Organization and project monthly hard spend limits are enforced (429 `project_spend_limit_exceeded`), with alert-only thresholds alongside. Enforcement isn't instantaneous, and there are no per-agent or per-session caps. ([source](https://developers.openai.com/api/docs/guides/spend-limits), checked 2026-11-06) | **Yes.** Per-agent and per-workspace spend budgets, model allowlists, and daily token limits. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Multi-provider models with fallbacks | **No.** Only OpenAI models (for example `gpt-6-astra`). You can switch the model for later turns (`POST /v1/agents/sessions/{id}`), but there is no multi-provider routing or automatic fallback. ([source](https://developers.openai.com/api/docs/guides/agents-api/configuration), checked 2026-11-06) | **Yes.** guild.yaml models is an ordered fallback list across Anthropic, OpenAI, Gemini, Meta, DeepSeek, and more; BYO keys supported. ([source](https://docs.guild.ai/guide/guild-yaml.md), checked 2026-10-23) |

### Observability

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Run traces | **Yes.** Tracing is on by default. Logs > Agents in the dashboard shows agent, generation and tool spans, and `GET /v1/agents/sessions/{id}/traces` exports OTLP JSON once the org enables trace export. ([source](https://developers.openai.com/api/docs/guides/agents-api/tracing), checked 2026-11-06) | **Yes.** Session event log of LLM calls, tool calls, sub-tasks, and container logs; also via the API. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Evaluations | **Partial.** Platform Evals, datasets and trace grading exist, but the agent-evals guidance targets Agents SDK and Agent Builder traces. No Agents API-specific eval harness is documented. ([source](https://developers.openai.com/api/docs/guides/agent-evals), checked 2026-11-06) | **Partial.** Evaluations with samples, checks, and llmJudge; enabled per account, no CLI command yet. ([source](https://docs.guild.ai/platform/evaluations.md), checked 2026-10-23) |
| Cost & usage tracking | **Yes.** Session and turn resources report best-effort token `usage` (input, cached, output, reasoning), split by root agent and subagent, and the dashboard shows usage too. These are not final bills, and there is no cache-write count. ([source](https://developers.openai.com/api/docs/guides/agents-api/observability), checked 2026-11-06) | **Yes.** Insights and Spend Explorer. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |

### Ecosystem & business

| Feature | OpenAI Agents API | Guild |
| --- | --- | --- |
| Public agent hub (install & fork) | **No.** There is no public hub for installing or forking Agents API agents. Codex plugin marketplaces and the ChatGPT plugin directory distribute plugins, not agents. ([source](https://learn.chatgpt.com/docs/developer-commands?surface=cli), checked 2026-11-06) | **Yes.** Agent Hub: Draft, Internal, or Public; owner~name with semver; install and fork (guild agent init --fork). ([source](https://docs.guild.ai/platform/publish-to-agent-hub.md), checked 2026-10-23) |
| Reusable skills | **Yes.** SKILL.md skills load from `environment.capability_directories`. Plugins (`.codex-plugin/plugin.json`) bundle skills and MCP config and can be uploaded as ZIPs to hosted sandboxes or reused through environment templates. ([source](https://developers.openai.com/api/docs/guides/tools-skills), checked 2026-11-06) | **Yes.** Hub skills, `guild skill`, and skills inside Goose and OpenClaw agents. ([source](https://docs.guild.ai/cli/commands.md), checked 2026-10-23) |
| Full lifecycle CLI | **Partial.** No Agents API lifecycle CLI is documented. The generic `openai` CLI covers Responses, files, images and admin, but its Agents commands are not documented. The Codex CLI (`codex cloud exec`, `codex exec-server`) covers Codex Cloud tasks and the self-hosted executor. ([source](https://developers.openai.com/api/docs/libraries/openai-cli), checked 2026-11-06) | **Yes.** 17 top-level commands: init, test, save --publish, workspace chat, trigger, integration, skill, session, and more. ([source](https://docs.guild.ai/cli/commands.md), checked 2026-10-23) |
| Open-source core | **Partial.** The Codex CLI, harness code, SDKs and app server are open source (Apache-2.0, github.com/openai/codex). The managed Agents API service is proprietary. ([source](https://learn.chatgpt.com/docs/open-source), checked 2026-11-06) | **No.** Proprietary platform; the CLI is published on npm. ([source](https://docs.guild.ai/llms.txt), checked 2026-10-23) |
| Public pricing | **Yes.** Public: there is no Agents API fee; model tokens and tools are billed at standard rates, and containers cost $0.03/$0.12/$0.48 per 20 minutes (1/4/16 GB), billed per minute with a 5-minute minimum. Auto-review tokens are extra. ([source](https://developers.openai.com/api/docs/pricing), checked 2026-11-06) | **Partial.** No public price list: plans are priced with the Guild team; model usage is metered at list price. ([source](https://docs.guild.ai/reference/pricing.md), checked 2026-10-23) |

## Frequently asked questions

### Is there a product called Codex managed agents?

Not by that name. OpenAI's managed agents built on Codex are the Agents API, which OpenAI describes as access to the Codex harness through an OpenAI-managed API. Codex Cloud is a separate ChatGPT product for coding tasks, and the Codex SDK drives local Codex threads.

### Can Guild agents use OpenAI models?

Yes. Guild agents choose their model in guild.yaml and can use OpenAI models alongside other providers, with fallbacks.

### How do I move an Agents API agent to Guild?

Copy the agent's instructions into a Native agent's PROMPT.md, or into an OpenClaw agent's AGENTS.md if it relies on Bash, files, and git. Replace each MCP server with a Guild integration, move vault credentials into Guild credentials, and recreate automations with guild trigger create. Skills keep the same SKILL.md format. Use the concept map above as the checklist.

### What doesn't map 1:1?

Self-hosted sandboxes (codex exec-server), steering a turn while it runs, and computer use. Guild runs agents in its own containers and starts them from triggers or chat.

### Is OpenAI Agents API pricing public?

Yes. There is no Agents API fee: you pay model tokens and tools at API rates, and hosted sandboxes cost $0.03, $0.12, or $0.48 per 20-minute session for 1, 4, or 16 GB.

## Sources

- [developers.openai.com/api/docs/guides/agents-api/overview](https://developers.openai.com/api/docs/guides/agents-api/overview)
- [developers.openai.com/api/docs/guides/agents-api/quickstart](https://developers.openai.com/api/docs/guides/agents-api/quickstart)
- [developers.openai.com/api/docs/guides/agent-builder](https://developers.openai.com/api/docs/guides/agent-builder)
- [developers.openai.com/api/docs/guides/agents](https://developers.openai.com/api/docs/guides/agents)
- [developers.openai.com/api/docs/guides/agents-api/sessions](https://developers.openai.com/api/docs/guides/agents-api/sessions)
- [developers.openai.com/api/docs/guides/agents-api/architecture](https://developers.openai.com/api/docs/guides/agents-api/architecture)
- [developers.openai.com/api/docs/guides/agents-api/multi-agent](https://developers.openai.com/api/docs/guides/agents-api/multi-agent)
- [developers.openai.com/api/docs/guides/agents-api/tools/mcp](https://developers.openai.com/api/docs/guides/agents-api/tools/mcp)
- [learn.chatgpt.com/docs/mcp-server](https://learn.chatgpt.com/docs/mcp-server)
- [developers.openai.com/api/docs/guides/agents-api/tools/functions](https://developers.openai.com/api/docs/guides/agents-api/tools/functions)
- [developers.openai.com/api/docs/guides/agents-api/tools/vaults](https://developers.openai.com/api/docs/guides/agents-api/tools/vaults)
- [developers.openai.com/api/docs/guides/agents-api/environments/self-hosted](https://developers.openai.com/api/docs/guides/agents-api/environments/self-hosted)
- [developers.openai.com/api/docs/guides/agents-api/environments/openai-hosted](https://developers.openai.com/api/docs/guides/agents-api/environments/openai-hosted)
- [developers.openai.com/api/docs/guides/agents-api/sessions/automations](https://developers.openai.com/api/docs/guides/agents-api/sessions/automations)
- [developers.openai.com/api/docs/guides/agents-api/sessions/webhooks](https://developers.openai.com/api/docs/guides/agents-api/sessions/webhooks)
- [developers.openai.com/api/docs/guides/agents-api/sessions/approvals](https://developers.openai.com/api/docs/guides/agents-api/sessions/approvals)
- [developers.openai.com/api/docs/guides/rbac](https://developers.openai.com/api/docs/guides/rbac)
- [developers.openai.com/api/docs/guides/admin-apis](https://developers.openai.com/api/docs/guides/admin-apis)
- [developers.openai.com/api/docs/guides/agents-api/environments/security](https://developers.openai.com/api/docs/guides/agents-api/environments/security)
- [developers.openai.com/api/docs/guides/spend-limits](https://developers.openai.com/api/docs/guides/spend-limits)
- [developers.openai.com/api/docs/guides/agents-api/configuration](https://developers.openai.com/api/docs/guides/agents-api/configuration)
- [developers.openai.com/api/docs/guides/agents-api/tracing](https://developers.openai.com/api/docs/guides/agents-api/tracing)
- [developers.openai.com/api/docs/guides/agent-evals](https://developers.openai.com/api/docs/guides/agent-evals)
- [developers.openai.com/api/docs/guides/agents-api/observability](https://developers.openai.com/api/docs/guides/agents-api/observability)
- [learn.chatgpt.com/docs/developer-commands](https://learn.chatgpt.com/docs/developer-commands?surface=cli)
- [developers.openai.com/api/docs/guides/tools-skills](https://developers.openai.com/api/docs/guides/tools-skills)
- [developers.openai.com/api/docs/libraries/openai-cli](https://developers.openai.com/api/docs/libraries/openai-cli)
- [learn.chatgpt.com/docs/open-source](https://learn.chatgpt.com/docs/open-source)
- [developers.openai.com/api/docs/pricing](https://developers.openai.com/api/docs/pricing)
- [docs.guild.ai/packages/agents-sdk](https://docs.guild.ai/packages/agents-sdk.md)
- [docs.guild.ai/guide/agent-types](https://docs.guild.ai/guide/agent-types.md)
- [docs.guild.ai/guide/guild-yaml](https://docs.guild.ai/guide/guild-yaml.md)
- [docs.guild.ai/integrations/overview](https://docs.guild.ai/integrations/overview.md)
- [docs.guild.ai/api-reference/mcp](https://docs.guild.ai/api-reference/mcp.md)
- [docs.guild.ai/platform/credentials](https://docs.guild.ai/platform/credentials.md)
- [docs.guild.ai/platform/credential-policies](https://docs.guild.ai/platform/credential-policies.md)
- [docs.guild.ai/platform/security-architecture](https://docs.guild.ai/platform/security-architecture.md)
- [docs.guild.ai/llms.txt](https://docs.guild.ai/llms.txt)
- [docs.guild.ai/platform/triggers](https://docs.guild.ai/platform/triggers.md)
- [guild.ai/product/ai-governance](https://www.guild.ai/product/ai-governance)
- [docs.guild.ai/platform/evaluations](https://docs.guild.ai/platform/evaluations.md)
- [docs.guild.ai/platform/publish-to-agent-hub](https://docs.guild.ai/platform/publish-to-agent-hub.md)
- [docs.guild.ai/cli/commands](https://docs.guild.ai/cli/commands.md)
- [docs.guild.ai/reference/pricing](https://docs.guild.ai/reference/pricing.md)

## Other comparisons

- [Guild vs. Claude Managed Agents](https://www.guild.ai/compare/claude-managed-agents)
- [Guild vs. AWS](https://www.guild.ai/compare/aws)
- [Guild vs. LangChain](https://www.guild.ai/compare/langchain)
- [Guild vs. Runlayer](https://www.guild.ai/compare/runlayer)
- [Guild vs. Arcade](https://www.guild.ai/compare/arcade)
- [Guild vs. Vercel eve](https://www.guild.ai/compare/vercel-eve)
- [Guild vs. CrewAI](https://www.guild.ai/compare/crewai)
- [Guild vs. Cursor Agent SDK](https://www.guild.ai/compare/cursor)
- [All comparisons](https://www.guild.ai/compare)
- [Book a demo](https://www.guild.ai/talk-to-sales)