简报·YT
← 返回 GitHub 动态
Significant-Gravitas/AutoGPT · 📈 热门项目 已完成

Significant-Gravitas/AutoGPT autogpt-platform-beta-v0.8.4

2026-10-14 16:22 · ⭐ 18.8万 · GitHub Release 原文 ↗
#开源AI #版本发布 #智能体平台

Significant-Gravitas/AutoGPT autogpt-platform-beta-v0.8.4

这个项目是什么

AutoGPT 是 AutoGPT Platform(原 AutoGPT 项目)的开源实现,目标是把"能自己拆解任务、调用工具、连续执行多步操作"的 AI 智能体做成一个可自托管、可订阅使用的平台。它值得关注的原因是:在大量 Agent 项目停留在 Demo 阶段时,这个项目仍在高频迭代真实的产品化能力——计费/订阅、MCP 工具接入、多模型路由、自托管部署、可观测性等,这次 v0.8.4 就是一个典型的"以修复和打磨为主"的版本。

一句话概括

本版本没有引入颠覆性新功能,主要围绕 Copilot 会话运行时与计费准确性、MCP 集成、自托管可用性、Sentry/PostHog 可观测性做了大量修复和体验优化。

要点

新增能力

修复的问题(重点)

Breaking Change

文档与其他

▸ Release Notes 英文原文(11167 字符)
# 🚀 Release `autogpt-platform-beta-v0.8.4` **Date:** October 2026 --- ## 🔥 What's New? ### New Features - **#15089** - Run Copilot chat on a turn runtime that resumes at its cursor on one connection (by @Pwuts) - **#15199** - Expert integrations connect through MCP first (by @0ubbe) - **#15248** - Add “Give feedback” to the account menu (hotfix) (by @kcze) - **#15216** - Keep Otto and expert replies short (by @0ubbe) - **#15217** - Make the expert hire card short and Typeform-like (by @0ubbe) - **#15249** - Send onboarding role to MailerLite and PostHog; exclude Iran and Russia from MailerLite (by @ntindle) - **#15204** - Redesign wallet with chat usage and credits (by @Torantulino) - **#15196** - Move onboarding paywall after personalization (by @Torantulino) - **#15061** - Approval cards label/word tool arguments from the tool schema (by @Pwuts) - **#15358** - Canceling a trial keeps access until the trial ends (by @itsababseh) - **#15361** - Sub-session waits for a slot when the user’s slots are full (by @Pwuts) - **#15043** - Approved call chain row expands to show what ran (by @Pwuts) - **#15349** - Add “Chats” to the collapsed sidebar (by @itsababseh) - **#15441** - Show “Thinking…” instead of backend status lines in Copilot (by @0ubbe) - **#15442** - Restyle chat markdown tables (by @0ubbe) - **#15462** - Give Copilot turns and graph node runs their own Sentry traces (by @ntindle) - **#15460** - Tag Sentry events with deployed commit and allow environment overrides (by @ntindle) - **#15213** - Route browser PostHog traffic through our own domain (by @kcze) - **#15480** - Add Meta Muse Wonder support (by @ntindle) - **#15466** - Send MailerLite segmentation fields to PostHog too (by @ntindle) - **#15496** - Add Google Gemini 4 Argon support (by @ntindle) - **#15498** - Attribute AutoPilot generations in PostHog and send reply feedback there (by @kcze) ### Enhancements - **#15461** - Add shared Sentry perf-compare recipe for performance agents (by @ntindle) ### UI/UX Improvements - **#15343** - Allow scrolling anywhere in the sidebar (by @Torantulino) - **#15407** - Show a single paywall toast with an upgrade link when a Copilot send is refused (by @ntindle) - **#15439** - Indicate no model is connected in builder chat; hide spend without billing (by @ntindle) - **#15448** - Prevent mobile expert pill overlap (by @0ubbe) - **#15455** - Line up mobile expert pill with the floating controls (by @0ubbe) ### Bug Fixes - **#15252** - Prevent long “Other” role input from losing onboarding profile (by @ntindle) - **#15250** - Use the right client for worker timezone and store checks (by @ntindle) - **#15309** - Install poetry and prisma in their own venvs so the image builds on debian:13-slim (by @Pwuts) - **#14148** - Index schedule reads by user and reduce jobstore queries per cache miss (by @ntindle) - **#15342** - Stop overcharging resumed SDK turns (hotfix) (by @ntindle) - **#15147** - Marketplace search finds experts by job title (by @Pwuts) - **#15149** - `find_capability` finds roster experts to hire and the user’s team (by @Pwuts) - **#15305** - Charge each SDK turn its own cost, not the session’s running total (by @Pwuts) - **#15182** - Bound AutoPilot trace export retries (by @ntindle) - **#15159** - Connect Telegram by messaging the bot first; adding to a group is optional (by @Bentlybro) - **#15254** - `bash_exec` kill-tree no longer leaves children alive during mid-listing exit (by @Bentlybro) - **#15346** - Start streaming a queued turn once backend runs it (no reload needed) (by @Pwuts) - **#15357** - PostHog flag reads carry email; dual mode logs only real disagreements (by @Pwuts) - **#15348** - Skip scheduled Copilot turn while owner is paywalled or over cost cap (by @Pwuts) - **#15153** - Library e2e: wait for search to apply before clicking a card (by @Pwuts) - **#15356** - Accept Web Push subscriptions from Edge, Safari, and pre-release Chrome (by @Pwuts) - **#15347** - Keep one running slot reserved for the user’s own message (by @Pwuts) - **#15353** - Queued messages and approval wakes start under their correct envelope (by @Pwuts) - **#15121** - Sync Codex transport tool rounds to the main turn limit (by @ntindle) - **#15370** - Load feature flags at startup in every service that reads them (by @Pwuts) - **#15380** - Show a lost send’s finished answer instead of “Failed to fetch” (by @Pwuts) - **#15382** - Plans bought during a canceled trial update the trial’s subscription (by @ntindle) - **#15381** - Retry Anthropic 5xx and overload errors reported by Claude CLI (by @ntindle) - **#15391** - Make Sentry reports readable and reduce instrumentation noise (by @ntindle) - **#15393** - Stop library, home, and AutoPilot pages waiting on the scheduler (by @ntindle) - **#15399** - Charge retried turns only for the attempt that succeeded (by @ntindle) - **#15310** - Strip only trailing `/api` in `getAGPTServerBaseUrl` (by @ntindle) - **#15311** - Route briefing footer choice via proxy; echo state on MCP OAuth errors (by @ntindle) - **#15331** - Stop `transformDates` rewriting user inputs and zone-less strings (by @ntindle) - **#15389** - Prevent per-user credentials lock from stalling Copilot for 60s (by @ntindle) - **#15395** - Make editing a schedule work end-to-end (by @ntindle) - **#15402** - Stale-tab message during another tab’s turn shows as queued follow-up (by @Pwuts) - **#15408** - Keep hosted-only UI out of self-hosted installs (by @ntindle) - **#15409** - Fix self-hosted signup grant, memory page, and no-model chat (by @ntindle) - **#15398** - Prevent false errors/spinners/endless retries after lost Copilot connection (by @Pwuts) - **#15410** - Stop sending a locked Advanced tier from a stored choice (by @Pwuts) - **#15413** - Replace a user’s credential list in a single transaction (by @ntindle) - **#15412** - Share subscription lookup cache across API workers (by @ntindle) - **#15420** - Transient retries bill once, never repeat tool calls, and work on first turns (by @ntindle) - **#15417** - Mask credentials in request URLs before logs and Sentry (by @ntindle) - **#15424** - Fix dropped-Escape test flakes and an unhovered usage tooltip (by @ntindle) - **#15425** - Log logout-closed WebSocket at debug, not failure (by @ntindle) - **#15426** - Create a new user profile on first read instead of returning 404 (by @ntindle) - **#15419** - Show which step failed and why when a run completes with errors (by @ntindle) - **#15066** - Sign in to Slack’s MCP server using a configured Slack app (by @Bentlybro) - **#15367** - HubSpot block: request service key/private app token instead of retired API key (by @Bentlybro) - **#15435** - Fail tests that leave DB transactions open; fix the offending test (by @ntindle) - **#15433** - Stop duplicate/noisy Sentry reports (by @ntindle) - **#15434** - Prevent optional select inputs crashing review cards and run forms (by @ntindle) - **#15438** - Name the missing field when a graph can’t be saved (by @ntindle) - **#15427** - Stop slow reads from hiding pending subscription changes (by @ntindle) - **#15366** - Sign in to HubSpot’s MCP server with configured app; refuse placeholder client where secret is required (by @Bentlybro) - **#15359** - Safely render .pptx and other Office files on Files page (SECRT-2437) (by @Abhi1992002) - **#15464** - Refresh Debian packages daily in image builds so security fixes ship (by @ntindle) - **#15467** - Read a trial user’s trial once per chat message (not four times) (by @ntindle) - **#15474** - Revert 1-hour SDK prompt cache (reverts #15470) (by @ntindle) - **#15481** - Label dev deployment as dev (not preview) in Sentry and PostHog (by @ntindle) - **#15369** - Provision managed Ayrshare profile when AutoPilot needs it (by @Bentlybro) - **#15482** - Make load tests authenticate through Better Auth (by @ntindle) - **#15484** - Keep an approved call’s result out of the user’s bubbles (by @Pwuts) - **#15221** - Hide trial offers for users with an existing plan (by @Torantulino) - **#15219** - Finish transactional email client compatibility (by @Torantulino) - **#14154** - Fix AI/ML API URL (blocks/llm) (by @hugoaimlapi) - **#14939** - Stop retransmitted chat send from queueing and running again (SECRT-2695) (by @Bentlybro) - **#15067** - Give every sidebar, popup, and builder chat the main chat’s cards (by @Bentlybro) - **#15128** - Run community Replicate models without a version hash (SECRT-2806) (by @Bentlybro) - **#15493** - Approval label reflects what each answered card sent to Otto (by @Pwuts) - **#15489** - Deleted shared runs stop serving their page and files (by @kcze) - **#15251** - Use “workflow” (not “agent”/“graph”) in user-facing copy (SECRT-2853) (by @itsababseh) ### Documentation - *(No documentation changes in this release)* ### Dependencies & Maintenance - **#15230** - Halve local test RAM and dependency image footprint (by @ntindle) - **#15354** - Drop leftover names of retired feature flags (by @Pwuts) - **#15360** - Keep OSS Scanner build and threat model in `.oss-scanner/` (by @ntindle) - **#15379** - Verify each chat route charges what its provider billed (by @ntindle) - **#15404** - Test billing audit mischarge report and clarify which relaunches keep the baseline (by @ntindle) - **#15397** - Add journeys leg to e2e for hosted-config user journeys (by @ntindle) - **#15405** - Frontend Copilot tests: assert UI after every way the connection can end (by @Pwuts) - **#15429** - Park compaction streams in tests instead of timing them (by @ntindle) - **#15428** - Pin retry shutdown boundary and compaction cut-short test coverage (by @ntindle) - **#15431** - Run AutoPilot journeys against a scripted model server (by @ntindle) - **#15072** - Bump dev dependencies (2 updates) (by @dependabot[bot]) - **#15158** - Bump production dependencies (2 updates) (by @dependabot[bot]) - **#15465** - Bump production dependencies (2 updates) (by @dependabot[bot]) - **#15457** - Apply Dependabot security bumps to backend dev dependencies (by @kcze) - **#15242** - Bump backend dev dependencies (7 updates) (by @dependabot[bot]) --- ## 🎉 Thanks to Our Contributors! A huge shoutout to everyone who contributed to this release. Special thanks to our new contributors: - @hugoaimlapi As well as our returning contributors: - @0ubbe - @Abhi1992002 - @Bentlybro - @dependabot[bot] - @itsababseh - @kcze - @ntindle - @Pwuts - @Torantulino --- ## 📥 How to Get This Update To update to this version, run: ```bash git pull origin autogpt-platform-beta-v0.8.4 ``` Or download it directly from the [Releases page](https://github.com/Significant-Gravitas/AutoGPT/releases). For a complete list of changes, see the [Full Changelog](https://github.com/Significant-Gravitas/AutoGPT/compare/autogpt-platform-beta-v0.8.3...autogpt-platform-beta-v0.8.4). --- ## 📝 **Feedback and Issues** If you encounter any issues or have suggestions, please [join our Discord](https://discord.gg/autogpt) and let us know!
▸ 主项目介绍 · Significant-Gravitas/AutoGPT
▸ README 原文(同步于 2026-07-19 13:27)
<p align="center"> <a href="https://platform.agpt.co/signup?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=hero_banner"> <img src="docs/home/.gitbook/assets/Banner_image.png" alt="AutoGPT" width="100%" /> </a> </p> <h1 align="center">AutoGPT — AI agents that finish the work</h1> <p align="center"> <strong>Get 10 hours back every week.</strong><br /> Describe what you want done. AutoGPT builds the agent, runs it, and reports back. </p> <p align="center"> <a href="https://platform.agpt.co/signup?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=header_get_started"><strong>Get started</strong></a> · <a href="https://platform.agpt.co/tour?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=tour_header">Tour</a> · <a href="https://agpt.co/pricing?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=header_pricing">Pricing</a> · <a href="https://docs.agpt.co">Docs</a> · <a href="https://discord.gg/autogpt">Discord</a> · <a href="#self-host-autogpt">Self-host</a> </p> <p align="center"> <a href="https://discord.gg/autogpt"> <img src="https://img.shields.io/badge/Discord-join-5865F2?logo=discord&logoColor=white" alt="Join Discord" /> </a> <a href="https://docs.agpt.co"> <img src="https://img.shields.io/badge/Docs-read-0A7AFF?logo=gitbook&logoColor=white" alt="Read the docs" /> </a> </p> --- ## The open-source platform for AI agents AutoGPT lets you build, deploy, and run AI agents that carry out complete workflows. Describe an outcome in plain English or shape every step in the visual builder, then run the agent on demand, on a schedule, or from a trigger. **185,000+ GitHub stars. Cited by:** | | | |---|---| | “Next frontier of prompt engineering imo: ‘AutoGPTs’.” | **Andrej Karpathy**, founding member of OpenAI | | “If you have a phone you can run AutoGPT. You don't even need to learn how to code.” | **Amjad Masad**, co-founder & CEO of Replit | | “AutoGPT might be the next big step in AI.” | **Lior Alexander**, CEO of AlphaSignal | --- ## Four surfaces, one platform <table> <tr> <td width="50%" align="center" valign="top"> <a href="https://agpt.co/product/autopilot/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=autopilot"> <img src="docs/content/imgs/readme/autogpt_autopilot_chat.jpg" alt="AutoPilot chat creating an AutoGPT agent" /> </a> <br /><strong><a href="https://agpt.co/product/autopilot/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=autopilot">AutoPilot</a></strong> <br />Describe the job in plain English and turn the conversation into a working agent. </td> <td width="50%" align="center" valign="top"> <a href="https://agpt.co/product/agents/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=agents"> <img src="docs/content/imgs/readme/autogpt_agent_dashboard.jpg" alt="Agents dashboard showing statuses, runs, and costs" /> </a> <br /><strong><a href="https://agpt.co/product/agents/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=agents">Agents</a></strong> <br />See every agent, run, cost, and action that needs your attention. </td> </tr> <tr> <td width="50%" align="center" valign="top"> <a href="https://agpt.co/product/marketplace/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=marketplace"> <img src="docs/content/imgs/readme/autogpt_marketplace.png" alt="AutoGPT Marketplace showing ready-made community agents" /> </a> <br /><strong><a href="https://agpt.co/product/marketplace/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=marketplace">Marketplace</a></strong> <br />Start from proven agents, add one to your library, and customize it for your work. </td> <td width="50%" align="center" valign="top"> <a href="https://agpt.co/product/build/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=build"> <img src="docs/content/imgs/readme/build_screen.jpg" alt="The AutoGPT Build canvas showing a real agent workflow" /> </a> <br /><strong><a href="https://agpt.co/product/build/?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=build">Build</a></strong> <br />Drag, connect, branch, and inspect blocks for exact control over every step. </td> </tr> </table> --- ## Get started ### AutoGPT Platform — public, hosted, and managed The hosted Platform is publicly available. We manage the infrastructure, model access, credentials, reliability, and updates so you can focus on the work your agents perform. **[Get started on AutoGPT Platform →](https://platform.agpt.co/signup?utm_source=github&utm_medium=referral&utm_campaign=autogpt_readme&utm_content=platform_get_started)** [Take the interactive tour →](https://platform.agpt.co/tour?utm_source=github&utm_medium=referral&utm_campaign=autogpt_readme&utm_content=tour_get_started) - AutoPilot, Agents, Marketplace, and Build - 45+ connected platforms and hundreds of AI models - No model API keys or infrastructure setup - Agents that run on demand, on schedules, and from triggers The hosted Platform is a paid service with usage-based agent runs. [Compare plans and pricing →](https://agpt.co/pricing?utm_source=github&utm_medium=referral&utm_campaign=autogpt_readme&utm_content=platform_pricing) ### Self-host AutoGPT > [!NOTE] > Self-hosting is the free path. You provide the infrastructure and model API keys, and you maintain the deployment. If you want zero setup, use the [managed Platform](https://platform.agpt.co/signup?utm_source=github&utm_medium=referral&utm_campaign=autogpt_readme&utm_content=self_host_note). **macOS and Linux:** ```bash curl -fsSL https://setup.agpt.co/install.sh -o install.sh && bash install.sh ``` **Windows PowerShell:** ```powershell powershell -c "iwr https://setup.agpt.co/install.bat -o install.bat; ./install.bat" ``` [Read the self-hosting guide →](https://docs.agpt.co/platform/getting-started) --- ## Managed Platform vs. self-hosting | | **AutoGPT Platform** | **Self-hosted** | |---|---|---| | Access | Public signup | Clone and install | | Cost | Paid plan plus agent usage | No license fee; pay your own infrastructure and model providers | | Setup | Managed | Docker and configuration required | | Model access | Built in | Bring your own API keys | | Updates and operations | Managed by AutoGPT | Managed by you | | Core builder and agent runtime | Included | Included | | Data and infrastructure control | Hosted by AutoGPT | Runs on your infrastructure | | Support | Plan-dependent | Community support | Both paths use the same repository. Choose the managed Platform when you want agents running immediately; self-host when infrastructure control matters more than operational convenience. --- ## Why the hosted Platform is paid Every agent run consumes real model usage, compute, storage, secrets management, and operational support. The managed Platform covers that infrastructure and funds continued development of the open-source project. Self-hosting remains available without a license fee for people and teams that want to provide and operate those resources themselves. --- ## What you can automate | Area | Example | |---|---| | **Executive operations** | Prepare a daily brief from internal and external signals | | **Sales** | Research every account before tomorrow's meetings | | **Marketing** | Turn a launch brief into campaign drafts across channels | | **Engineering** | Triage incidents and start with a likely cause | | **Customer support** | Draft replies, collect context, and flag escalations | | **Research** | Monitor sources and return structured reports when something changes | --- ## Integrations Connect the apps that are yours. AutoGPT provides access to hundreds of AI models and connects agents to 45+ platforms, including: `Gmail` · `Google Calendar` · `Google Docs` · `Google Sheets` · `GitHub` · `Slack` · `Discord` · `Notion` · `HubSpot` · `Linear` · `Airtable` · `Jira` · `Salesforce` · `Stripe` · `Webflow` [Explore the integrations →](https://agpt.co/docs/integrations) --- ## Community and support | Resource | Link | |---|---| | Discord | [Join the AutoGPT community](https://discord.gg/autogpt) | | Documentation | [docs.agpt.co](https://docs.agpt.co) | | Bug reports | [GitHub Issues](https://github.com/Significant-Gravitas/AutoGPT/issues/new/choose) | | Feature requests | [GitHub Discussions](https://github.com/Significant-Gravitas/AutoGPT/discussions) | | Contributing | [CONTRIBUTING.md](CONTRIBUTING.md) | --- ## License | Component | License | What it means | |---|---|---| | `autogpt_platform/` | [Polyform Shield](https://polyformproject.org/licenses/shield/1.0.0/) | Free for personal and internal business use; cannot be sold as a competing hosted service | | `classic/` and everything else | [MIT](LICENSE) | Permissive open-source use | --- ## AutoGPT Classic Looking for the original standalone AutoGPT agent? It remains available in [`classic/`](classic/) under the MIT License. - [Build an agent with Forge](classic/FORGE-QUICKSTART.md) - [Benchmark an agent with `agbenchmark`](https://pypi.org/project/agbenchmark/) - [Explore the Classic project](classic/README.md) --- ## Contributors <a href="https://github.com/Significant-Gravitas/AutoGPT/graphs/contributors"> <img src="https://contrib.rocks/image?repo=Significant-Gravitas/AutoGPT&max=1000&columns=10" alt="AutoGPT contributors" /> </a> <p align="center"> <a href="https://platform.agpt.co/signup?utm_source=github&amp;utm_medium=referral&amp;utm_campaign=autogpt_readme&amp;utm_content=footer_get_started"><strong>Get started with AutoGPT →</strong></a> </p> --- <!-- Keep these links. Translations will automatically update with the README. --> [Deutsch](https://zdoc.app/de/Significant-Gravitas/AutoGPT) | [Español](https://zdoc.app/es/Significant-Gravitas/AutoGPT) | [français](https://zdoc.app/fr/Significant-Gravitas/AutoGPT) | [日本語](https://zdoc.app/ja/Significant-Gravitas/AutoGPT) | [한국어](https://zdoc.app/ko/Significant-Gravitas/AutoGPT) | [Português](https://zdoc.app/pt/Significant-Gravitas/AutoGPT) | [Русский](https://zdoc.app/ru/Significant-Gravitas/AutoGPT) | [中文](https://zdoc.app/zh/Significant-Gravitas/AutoGPT)