GitHub - demeyer1/Autobot: Autobot is a privacy-aware operating harness that integrates via native ChatGPT Project capabilities integrating OpenAI synchronous voice and OAI UI and control planes. This is an internal, knowledge worker tool leveraged by the team at Autonomous Production tool · GitHub
Skip to content
Navigation MenuSign inAppearance settingsPlatformAI CODE CREATIONGitHub CopilotWrite better code with AIGitHub Copilot appDirect agents from issue to mergeMCP RegistryIntegrate external toolsDEVELOPER WORKFLOWSActionsAutomate any workflowCodespacesInstant dev environmentsIssuesPlan and track workCode ReviewManage code changesCode QualityEnforce quality at mergeAPPLICATION SECURITYGitHub Advanced SecurityFind and fix vulnerabilitiesCode securitySecure your code as you buildSecret protectionStop leaks before they startEXPLOREWhy GitHubDocumentationBlogChangelogMarketplaceView all featuresSolutionsBY COMPANY SIZEEnterprisesSmall and medium teamsStartupsNonprofitsBY USE CASEApp ModernizationDevSecOpsDevOpsCI/CDView all use casesBY INDUSTRYHealthcareFinancial servicesManufacturingGovernmentView all industriesView all solutionsResourcesEXPLORE BY TOPICAISoftware DevelopmentDevOpsSecurityView all topicsEXPLORE BY TYPECustomer storiesEvents & webinarsEbooks & reportsBusiness insightsGitHub SkillsSUPPORT & SERVICESDocumentationCustomer supportCommunity forumTrust centerPartnersView all resourcesOpen SourceCOMMUNITYGitHub SponsorsFund open source developersPROGRAMSSecurity LabMaintainer CommunityGitHub StarsArchive ProgramREPOSITORIESTopicsTrendingCollectionsEnterpriseENTERPRISE SOLUTIONSEnterprise platformAI-powered developer platformAVAILABLE ADD-ONSGitHub Advanced SecurityEnterprise-grade security featuresCopilot for BusinessEnterprise-grade AI featuresPremium SupportEnterprise-grade 24/7 supportPricingSearch/Sign inSign upAppearance settings
You signed in with another tab or window. Reload to refresh your session.
You signed out in another tab or window. Reload to refresh your session.
You switched accounts on another tab or window. Reload to refresh your session.
Dismiss alert
demeyer1
/
Autobot
Public
Notifications
You must be signed in to change notification settings
Fork
1
Star
8
Code
Issues
0
Pull requests
0
Actions
Projects
Security and quality
0
Insights
Additional navigation options
Code
Issues
Pull requests
Actions
Projects
Security and quality
Insights
mainBranchesTagsGo to fileCodeOpen more actions menuLatest commit History65 Commits65 CommitsFolders and filesNameNameLast commit messageLast commit date00_CONTEXT00_CONTEXT 01_PROJECTS/_template01_PROJECTS/_template 02_INBOX02_INBOX 03_OUTPUTS03_OUTPUTS benchmarksbenchmarks configconfig docsdocs runtimeruntime scriptsscripts skills/first-timeskills/first-time teststests .gitignore.gitignore AGENTS.mdAGENTS.md CHANGELOG.mdCHANGELOG.md INSTALL_FOR_AI.mdINSTALL_FOR_AI.md Install.commandInstall.command LICENSELICENSE PRIVACY.mdPRIVACY.md PROJECTS.mdPROJECTS.md README.mdREADME.md SECURITY.mdSECURITY.md VERSIONVERSION install.shinstall.sh uninstall.shuninstall.sh View all filesRepository files navigationREADMEMIT licenseSecurityMore itemsAutobot
AutoBot: a self-improving agentic harness that makes frontier AI better at finishing complex knowledge work.
AutoBot achieved 18.5% higher task completion than the published OpenAI Sol Max baseline, surpassing Anthropic’s Claude Opus 5 Max on OSWorld 2.0, a benchmark of long, multi-application workflows. It also reached #1 on the official AssistantBench hidden-test leaderboard. Results and methodology
Hard workflows become upgrades to the agent itself. AutoBot repairs its own harness, independently validates the changes, and carries them forward. The next workflow inherits the improvement. Compounding capability, without retraining the model.
Your knowledge outgrows the context window. Hierarchical memory lives on disk; task-specific retrieval builds the working context. Nightly consolidation integrates new knowledge and corrections. Your agent accumulates institutional memory across projects and conversations.
Your project can outlive the agent working on it. Persistent task graphs, atomic checkpoints and independent supervision let a replacement worker resume the assignment. Completion is bound to current requirements and verified destination evidence.
Local compute makes persistent intelligence economical. Your CPU handles orchestration, state and integrity checks. Compiled context and reusable proofs reduce repeated inference, directing the model’s budget toward the difficult judgments that move work forward.
Open source. Native ChatGPT on your Mac. Built by Autonomous Production. Get AutoBot.
Benchmarks
AssistantBench
#1 on the recorded official hidden-test leaderboard: 50.70% accuracy across 181 tasks.
AutoAssist was the harness's previous brand name, changed to AutoBot two weeks after this screenshot.
Results and verification
OSWorld 2.0
32.41% binary accuracy and 64.28% partial accuracy across 108 tasks. Final best-valid-per-task aggregate, compared with the published September 10, 2026 leaderboard snapshot.
Results and methodology
What it changes
Privacy by relationship and purpose. Personal, family and friends, work, and deliberately shared context have separate zones. Nothing moves into SHARED automatically.
Independent completion. The local runtime rejects validation under the producer label. The operating contract also requires a separate validator to inspect current evidence, including the real destination when the task changes something outside the workspace.
Durable follow-through. Native ChatGPT Goals keep the active work moving. Autobot keeps the objective, stages, evidence, and recovery state on disk so an interrupted chat does not silently erase the commitment.
Clean external actions. Recipient-visible writes use the signed-in first-party app through Computer Use. The operating contract requires the active ChatGPT workflow to verify the account, destination, and visible content before the action, then check the rendered result for a duplicate, failure, or unwanted AI attribution. If the clean route cannot be verified, the workflow must stop.
Tone that stays in its lane. Communication profiles are separated by channel and audience. A family text profile does not become a work email profile. Autobot stores compact, user-approved patterns rather than raw message archives by default.
What Autobot includes
Autobot is the combination of two layers:
Native ChatGPT: the desktop app, local Projects, Voice, Goals, skills and plugins, scheduled work, notifications, and Computer Use.
The Autobot workspace: the operating contract in AGENTS.md, privacy zones, selective memory, project status, a local objective state machine, a one-minute liveness supervisor, external-action policy, and first-time setup.
Autobot is not a separate model, chatbot, or agent gateway. It depends on current ChatGPT capabilities and their plan, region, usage, sandbox, and permission limits. See Architecture and Permissions.
Autobot compared with OpenClaw and Hermes
This comparison evaluates Autobot plus native ChatGPT for one person managing personal and work tasks on a Mac. "Better" means a more explicit default for the stated user need, based on the published design; it does not mean a measured advantage in accuracy, speed, or reliability. "Doesn't do" means the specific built-in requirement was not found in the primary documentation reviewed on September 10, 2026. Both alternatives can be extended.
Top 3 shared capabilities: Autobot's approach and user benefits
Top 3 additional Autobot capabilities and user benefits
OpenClaw
1. Remember context with explicit boundaries. OpenClaw persists and searches memory. Autobot adds default personal, family/friends, work, and opt-in shared zones. This makes the rules for reusing private context in a work task more explicit. Memory / Autobot privacy.2. Track the promised outcome. OpenClaw records background tasks and delivery state. Autobot assigns each promised output an owner, destination, and completion gate. This gives the user a clearer record of what remains owed after an interruption. Tasks / Autobot contract.3. Check the result of an app action. OpenClaw supports tools and outbound audit history. Autobot requires account, destination, and content checks before a write, followed by rendered inspection. This adds an explicit check for a wrong destination, failed save, or duplicate action. Audit / Autobot writes.
1. Require a separate completion validator. Autobot's five-stage process requires current evidence and a validator label different from the producer, with destination readback for external work. This gives users evidence beyond the worker's own success report. Completion.2. Block writes when clean delivery cannot be verified. Autobot's default policy requires first-party Computer Use and stops a route that forces unwanted attribution. This gives users an explicit publication rule instead of relying on each connector's behavior. Write policy.3. Keep learned tone separate by channel and audience. Autobot requires authorized examples and compact, separate communication profiles. This helps keep a family-text style from shaping a customer email. Profiles.
Hermes
1. Remember context with purpose-specific retrieval. Hermes has curated memory, user profiles, and session search. Autobot makes relationship and purpose part of its default retrieval rules. This gives users clearer control over which personal facts may inform professional work. Memory / Autobot privacy.2. Follow work through to its deliverable. Hermes schedules jobs and delivers their outputs. Autobot also retains the objective, ordered evidence stages, and unfinished outputs. This makes it easier to distinguish a job that ran from a requested artifact that was saved and checked. Scheduling / Autobot completion.3. Verify user-visible changes. Hermes provides tool approvals and execution controls. Autobot adds a standard before-and-after inspection of the signed-in destination. This gives the user a specific check that an authorized action produced the intended visible result. Security / Autobot writes.
1. Require independent acceptance of every completion stage. Autobot separates the producer and validator labels and requires fresh destination evidence for external outputs. This makes an unsupported "done" report insufficient under the operating contract. Completion.2. Require a clean first-party write route by default. Autobot checks the exact account and visible content, then rejects forced-attribution or unverifiable delivery routes. This gives users a consistent rule for communications across services. Write policy.3. Require audience-specific tone-learning boundaries. Hermes supports personality and user-style preferences; Autobot specifies separately authorized profiles for each channel and audience. This gives users more explicit control over which examples shape each kind of message. Hermes memory / Autobot profiles.
These are operating-contract differences, not guarantees of error-free execution. Autobot's privacy zones and validator separation are procedural within one Mac; they are not OS isolation or cryptographic identities. OpenClaw and Hermes offer broader standalone deployment, messaging, and model choices. The absence findings above concern the exact required workflows, not an absence of memory, privacy controls, voice, verification tools, or automation in either project.
What is new in 0.3.0
The project-local first-time skill now acts as the setup operator. It checks current state, applies safe local defaults, runs supported actions, preserves the earliest unresolved user or capability gate, and returns to the user's original task only after current marker and status readback.
The base folder installs without Node or sign-in. The advanced runtime needs Node 22 or newer. Native app capabilities are discovered separately; no model, account, notification recipient or external write is configured for you. See capabilities and availability.
Privacy zones
Zone
Intended context
Default boundary
PRIVATE
Sensitive personal facts, preferences, health, finances, and private plans
Never shared automatically
FAMILY_FRIENDS
Relationships, events, logistics, and authorized personal communication patterns
Never used for work without a current explicit need
WORK
Organizations, projects, teammates, customers, vendors, and professional communication patterns
Never used for personal communication unless explicitly relevant
SHARED
The minimum facts you deliberately make reusable across zones
Opt-in only, with provenance
These are owner-only folders and operating rules inside one macOS account. They are not separate encrypted vaults, macOS users, or hardware security boundaries. Use separate macOS accounts or separate machines when the trust domains require stronger isolation. Read Privacy.
Completion that requires evidence
Each durable objective moves through five ordered stages:
research_complete
draft_complete
destination_updated
save_confirmed
rendered_readback_verified
The local runtime enforces stage order, evidence hashes, and a validator label different from the producer label. The operating contract requires a genuinely separate validator to use fresh evidence and inspect the authoritative destination for external work.
This is procedural independence on one Mac, not a separate security enclave. Autobot cannot bypass a login, MFA, macOS permission, plan limit, user decision, or the scope of the user's instruction. The supervisor records liveness and flags stalled objectives. It does not control desktop apps in the background or manufacture permission to continue.
No-attribution, fail-closed writes
Autobot treats connectors, apps, MCP tools, browser integrations, and APIs as read-only unless a destination-specific adapter proves the same clean-write properties. Recipient-visible writes normally use the signed-in first-party interface through native ChatGPT Computer Use.
Before a live write, the operating contract requires the active ChatGPT workflow to check the app, account, destination, scope, and final visible content. Afterward, the workflow must inspect the rendered result and check for the intended mutation, no duplicate, no failure, and no added AI or ChatGPT attribution. If a platform forces a non-removable label or the result is ambiguous, the workflow must not claim success.
That policy cannot remove immutable metadata or disclosures controlled by a third-party platform. Details: Security.
Native Voice and long-running work
ChatGPT Voice can start separate threads for longer tasks, check them, send follow-ups, and bring progress or blockers back into the voice conversation. Goal mode keeps the outcome and completion criteria attached to the work.
Important limits:
Voice availability, usage, and rollout depend on the ChatGPT plan and workspace.
Only one voice chat can be active across desktop at a time.
Tasks started from Voice also use the Codex usage budget.
Local long-running work needs the Mac and workspace to remain available. Enable Prevent sleep while running when appropriate.
Starting a Goal does not broaden sandbox access or approval authority.
Commands
./runtime/bin/autoassist doctor
./runtime/bin/autoassist version
./runtime/bin/autoassist core status
./runtime/bin/autoassist core help
Run ./runtime/bin/autoassist help for the complete local command list.
privacy-scan is a clean release-candidate check, not a post-install scan of a populated user workspace. Run it only before user configuration or against a separate clean release tree.
Read next
Install
Permissions
Capabilities and availability
Architecture
Local runtime reference
Privacy
Security
Autobot and OpenClaw
Troubleshooting
Current status
Autobot 0.3.0 is an early public release for a single user on a Mac. The package is not a multi-tenant service, an OS sandbox, a cryptographic privacy boundary, or a guarantee that every third-party action will succeed. Treat a downloadable ZIP as verified only when its checksum and manifest match the adjacent release artifacts and the packaged candidate passes the bundled release checks.
License
Autobot is available under the MIT License. Copyright (c) 2026 Autobot contributors.
Setup and installation
Set up AutoBot on your MacBook, then use it from your phone.
1 Get the Mac app for Codex
On your MacBook, download OpenAI's desktop app, sign in, and select Codex. The download is called ChatGPT, with Codex inside it.
2 Choose your plan
AutoBot is free, but regular use will likely need a paid ChatGPT plan; choose one that fits your needs.
3 Let it use your Mac
In Plugins > Computer Use, choose Install or Enable and turn on its server and skill switches. In your Mac's System Settings > Privacy & Security, enable the named OpenAI app or helper for:
Screen Recording or Screen & System Audio Recording to see your screen.
Accessibility to click and type.
Microphone to talk with you.
Automation, when requested, to work with the named app.
Files and Folders, when requested, to use the folders you select.
Stay at your MacBook and plan for at least three permission-related app restarts during first setup; the number varies. Reopen the app after each prompt and continue.
4 Set your everyday preferences
Open Settings with Command + comma and choose these preferences:
General > Prevent sleep while running: On. Keep the MacBook plugged in, online, and open for remote work.
Notifications: On for finished tasks and questions.
General > Permissions: enable your choice, then select it below the message box. Full access allows broad file and internet actions; Approve for me (Auto-review) reviews requests automatically; Ask for approval brings requests to you.
Computer Use: choose your apps; Always allow lets it reuse them.
Model menu: try Sol, Max reasoning, and Fast off; change these anytime.
5 Download AutoBot
Open the AutoBot release and download AutoAssist-v0.3.0.zip plus AutoAssist-v0.3.0.zip.sha256. Keep them together and double-click the ZIP to open the download.
6 Start inside a project
Create a local project called AutoBot in Codex, choose the downloaded AutoAssist folder, and start a task there. Paste: “Install AutoBot from https://github.com/demeyer1/Autobot using INSTALL_FOR_AI.md, including required support software from official sources, then guide setup.”
7 Point the project at your installed AutoBot
After installation, choose Edit project > Add folder, select AutoAssist in your Mac's home folder, and choose Make primary. Start a fresh task there: “Run first-time setup for this installed folder.”
8 Connect your phone
Install the ChatGPT mobile app, sign in to the same account and workspace, and on your Mac open Settings > Connections > Control this Mac or PC > Set up. Scan the QR code, finish verification, and open Remote on your phone.
9 Add Chrome
Open Settings > Computer Use > Chrome > Install and add the ChatGPT extension to your chosen Chrome profile. Return to settings and check for Manage.
10 Connect your everyday apps
Tell AutoBot, “Connect and test my email, calendar, Slack, and iMessage,” then sign in to your chosen accounts, including Messages on your Mac. Use available plugins for reading and Computer Use in the signed-in apps for sending.
11 Try your first task
On your phone, open Remote > AutoBot and ask, “Make and save tomorrow's to-do list.” Open the result, then try voice.
Three tips and tricks
Give it a useful Chrome profile. Choose one with the logins, passwords, and autofill you want it to use, and ask it to connect your usual apps.
Teach it how you sound. Share a few example messages, correct its drafts, and save separate styles for customers, teammates, and friends.
Explain your boundaries. Explain what belongs to work, personal life, or both, which accounts to use, and when each tone applies.
Privacy and safety
Keep sensitive work in your hands. Require your approval for private or sensitive access, sharing, and changes, plus a standalone third-party authenticator app on a separate device AutoBot cannot access, mirror, or sync; SMS or iMessage alone doesn't meet that boundary.
Make the challenge real. A saved login can bypass a new sign-in challenge, and AutoBot doesn't add one to every action; use services that require fresh verification for sensitive actions, or keep those accounts disconnected and handle them yourself.
Choose what stays separate. Keep highly sensitive files and accounts outside AutoBot's connected folders and Chrome profile, and store recovery codes on the separate device or offline.
Choose where your data goes. Review ChatGPT's data controls and connected services before adding private material, and keep a backup of your workspace; a local folder doesn't mean every request stays on your Mac.
If you need a hand later
Setup paused: reopen the same project and say, “Continue first-time setup from where we left off.”
Something isn't working: say, “Check my AutoBot setup and help me fix what's missing.”
Remote is missing: update both apps and check that your phone and Mac use the same account and workspace.
Update or remove it: say, “Help me update AutoBot,” or “Remove AutoBot and keep my files recoverable.”
Setup references: Mac permissions, permission choices, phone pairing, and MFA guidance.
For ChatGPT: automated installation
If the user asks you to install this repository, read INSTALL_FOR_AI.md first. It contains the commands, local project setup, verification and resume instructions.
The user can start with: “Go to https://github.com/demeyer1/Autobot and install AutoBot in a new local project, following INSTALL_FOR_AI.md.”
AboutAutobot is a privacy-aware operating harness that integrates via native ChatGPT Project capabilities integrating OpenAI synchronous voice and OAI UI and control planes. This is an internal, knowledge worker tool leveraged by the team at Autonomous Production tool autoprod.aiResourcesReadmeMIT licenseSecurity policySecurity policyActivityStars8 starsWatchers0 watchingForks1 forkReport repositoryReleasesPackagesContributorsLanguages
Footer
© 2026 GitHub, Inc.
Footer navigation
Terms
Privacy
Security
Status
Community
Docs
Contact
Manage cookies
Do not share my personal information
You can’t perform that action at this time. |
Autobot is introduced as a privacy-aware operating harness designed to enhance frontier AI capabilities for complex knowledge work by integrating native ChatGPT Project features with OpenAI synchronous voice and OAI UI and control planes. It functions as a self-improving agentic system that aims to elevate AI performance through iterative refinement of its operational structure. The core methodology centers on allowing the agent to repair its own harness, independently validate proposed changes, and carry those improvements forward, enabling a compounding capability without requiring full model retraining.
The system addresses the limitations of context windows by utilizing hierarchical memory stored on disk for longer-term knowledge, while task-specific retrieval is employed to build the working context for immediate operations. The agent accumulates institutional memory across various projects and conversations through nightly consolidation, establishing a persistent knowledge base. Furthermore, it supports the persistence of tasks through task graphs, atomic checkpoints, and independent supervision, allowing a replacement worker to seamlessly resume assignments, bound by verified destination evidence and current requirements. This entire operational framework leverages local compute to manage orchestration, state, and integrity checks economically, directing the model’s resources toward complex judgments that advance the work.
A critical aspect of Autobot’s design is its commitment to privacy managed through explicit context separation. The system establishes distinct privacy zones: PRIVATE, FAMILY_FRIENDS, WORK, and SHARED, which dictate how personal, professional, and shared contexts interact, ensuring that sensitive information does not automatically migrate between zones. This adherence to boundaries is formalized within an operating contract that governs how context is handled and shared.
The execution of tasks incorporates rigorous verification protocols. The system features a multi-stage completion process, including research_complete, draft_complete, destination_updated, save_confirmed, and rendered_readback_verified. The local runtime enforces the required order of these stages, utilizes evidence hashes, and requires a separate validator label distinct from the producer label to inspect the authoritative destination before external actions are taken. This procedural independence is crucial, as the system requires a genuinely separate validator to inspect current evidence and verify the destination for external work, ensuring that actions are verified against real-world outcomes.
Regarding external actions, Autobot implements a policy that blocks writes if a clean delivery route cannot be verified. It mandates that recipient-visible writes use the signed-in first-party application through the native ChatGPT Computer Use feature. Before any live write, the workflow must verify the application, account, destination, and scope, and subsequently inspect the rendered result to confirm the intended mutation, check for errors or duplicates, and ensure no unwanted AI attribution is present. If this verification fails, the workflow must halt. This policy ensures a consistent and explicit publication rule.
The system also manages communication tone by separating communication profiles based on channel and audience, preventing contexts like family text from influencing professional email profiles. This is achieved by storing compact, user-approved patterns rather than raw message archives by default.
In comparative analyses with other systems, such as OpenClaw and Hermes, Autobot introduces distinct operational contracts. For instance, regarding memory, Autobot mandates explicit boundaries for context reuse across personal and work domains, assigning owners and completion gates to track promised outcomes. Furthermore, concerning external actions, Autobot requires independent acceptance of every completion stage and demands a clean first-party write route, which provides an explicit safeguard against forced attribution or unverifiable delivery routes. Similarly, it requires audience-specific tone-learning boundaries, offering greater control over how personalization shapes different communication styles.
The installation and setup process involves configuring necessary permissions, enabling system features for screen recording, accessibility, microphone access, and file/folder usage for the necessary external applications, all managed through the macOS System Settings. The system requires specific user preferences, such as enabling prevention of sleep while running and managing notification settings. The overall architecture depends on native ChatGPT capabilities, and limitations exist concerning voice availability, usage quotas, and the need for continuous availability of the local machine and workspace for long-running tasks. For deployment, the system is offered under the MIT License, and installation hinges on following specific commands and documentation to set up the local runtime and connect the system to the installed OpenAI application. |