# agentsite-static-version=991f9c27-98d6-4cef-b279-198cda58d54d
# pneum.ai

> AI-optimized mirror of pneum.ai containing 50 pages totalling 16,602 words of clean markdown content, structured data, and semantic HTML. Original source: https://pneum.ai. Last updated: 2026-09-16T03:03:51.285Z.

Each page is available as HTML (with JSON-LD structured data) and Markdown (text-only, ideal for LLMs and RAG).

## Homepage

- [AI Assistant That Gets Things Done | Pneum.ai](/content/site-root.html): Talk or type to an assistant that can research, work across connected services, control supported devices, and ask before important changes. (491 words)

## Articles & Blog Posts

- [Start Pro | Pneum.ai](/content/start/pro/index.html): Confirm your Pneum.ai pro plan before creating an account. (138 words)
- [Install and manage integrations | Pneum.ai Docs](/content/docs/integrations/overview/index.html): Browse the marketplace, select tools, connect credentials, set boundaries, and verify an integration safely. (354 words)
- [Launch field reference | Pneum.ai Docs](/content/docs/reference/launch-fields/index.html): Use a compact field-by-field reference for the Cloud agent launch dialog. (266 words)
- [Start Basic | Pneum.ai](/content/start/basic/index.html): Confirm your Pneum.ai basic plan before creating an account. (232 words)
- [Glossary | Pneum.ai Docs](/content/docs/reference/glossary/index.html): Definitions for Cloud agents, runtimes, profiles, tools, integrations, models, approvals, and billing terms. (337 words)
- [Troubleshoot connection and voice | Pneum.ai Docs](/content/docs/troubleshooting/connection-and-voice/index.html): Diagnose unreachable public or tailnet endpoints, runtime readiness, microphones, transcription, speech, and Voice Studio. (332 words)
- [Network and permission boundaries | Pneum.ai Docs](/content/docs/security/network-boundaries/index.html): Combine cloud exposure, firewall rules, connector allowlists, tool profiles, and approvals into layered access control. (312 words)
- [Agent status reference | Pneum.ai Docs](/content/docs/reference/agent-statuses/index.html): Interpret every Cloud agent badge and know which lifecycle actions are safe in each state. (276 words)
- [Troubleshoot integrations and billing | Pneum.ai Docs](/content/docs/troubleshooting/integrations-and-billing/index.html): Diagnose connector authentication and allowlists, then distinguish estimates, synchronized usage, and invoice issues. (306 words)
- [Troubleshooting overview | Pneum.ai Docs](/content/docs/troubleshooting/overview/index.html): Collect useful evidence, identify the failing layer, and choose the right recovery guide without making the problem worse. (241 words)
- [Plan limits reference | Pneum.ai Docs](/content/docs/reference/plan-limits/index.html): Compare agent, CPU, GPU, managed-model, open-model, and custom-voice limits by plan. (221 words)
- [Security | Pneum.ai](/content/security/index.html): Learn how Pneum.ai isolates assistants, encrypts connector credentials, controls connected actions, handles customer content, and deletes accounts. (341 words)
- [Troubleshoot launch and agent status | Pneum.ai Docs](/content/docs/troubleshooting/launch-and-status/index.html): Recover from unavailable options, failed provisioning, long transitions, start or stop errors, and cleanup issues. (342 words)
- [Connect a Tailscale tailnet | Pneum.ai Docs](/content/docs/security/tailscale/index.html): Configure one account-level OAuth connection for private agent launches, DNS, routes, and custom tools. (326 words)
- [HTTP, MCP, and container custom tools | Pneum.ai Docs](/content/docs/integrations/custom-tools/index.html): Define a custom tool schema, configure its runtime safely, test it, and enable it under an approval policy. (364 words)
- [Select a coding workspace | Pneum.ai Docs](/content/docs/coding/workspaces/index.html): Give coding agents a bounded folder, understand container isolation, and keep repository credentials outside the sandbox. (289 words)
- [Microsoft 365 and Teams | Pneum.ai Docs](/content/docs/integrations/microsoft-365/index.html): Connect Outlook, calendar, OneDrive, SharePoint, Teams tools, and the optional Teams chat interface. (451 words)
- [Test and deploy coding changes | Pneum.ai Docs](/content/docs/coding/testing-and-deployment/index.html): Run focused checks, preview web applications, and publish reviewed changes without bypassing approvals. (278 words)
- [Slack | Pneum.ai Docs](/content/docs/integrations/slack/index.html): Configure Slack tools and the Socket Mode chat interface with channel, DM, mention, and approval controls. (330 words)
- [Run coding work in background agents | Pneum.ai Docs](/content/docs/coding/background-tasks/index.html): Delegate a bounded coding objective, monitor progress, steer safely, and review the resulting workspace changes. (249 words)
- [Telegram | Pneum.ai Docs](/content/docs/integrations/telegram/index.html): Configure a BotFather bot for allowed chats, proactive tools, inbound chat, audio, commands, and approvals. (269 words)
- [Jira Cloud | Pneum.ai Docs](/content/docs/integrations/jira/index.html): Connect Jira with an account token, restrict access by project key, and enable issue workflow actions. (288 words)
- [Discord | Pneum.ai Docs](/content/docs/integrations/discord/index.html): Configure a Discord bot token, channel IDs, Message Content intent, inbound chat policy, and tools. (290 words)
- [Salesforce | Pneum.ai Docs](/content/docs/integrations/salesforce/index.html): Connect a Salesforce session, allow specific object types, and control record reads and writes. (257 words)
- [Philips Hue | Pneum.ai Docs](/content/docs/integrations/philips-hue/index.html): Connect a Hue Bridge over the local network, create an application key, discover lights, and enable controls. (267 words)
- [GitHub | Pneum.ai Docs](/content/docs/integrations/github/index.html): Connect repositories for governed reads and writes, then publish coding changes through immutable bundles. (313 words)
- [Tools Studio and tool profiles | Pneum.ai Docs](/content/docs/studios/tools/index.html): Control which tools background agents can use, attach skills, and define approval behavior with least privilege. (293 words)
- [Documentation | Pneum.ai](/content/docs/index.html): Learn how to launch, configure, operate, secure, and extend Pneum.ai Cloud agents. (168 words)
- [Approvals, permissions, and steering | Pneum.ai Docs](/content/docs/agents/approvals-and-steering/index.html): Review tool arguments and diffs, choose the narrowest approval, and redirect active or channel work safely. (351 words)
- [Launch options reference | Pneum.ai Docs](/content/docs/cloud/launch-options/index.html): Choose a provider, runtime, region, instance, disk, exposure mode, and model credential strategy. (515 words)
- [Pricing | Pneum.ai](/content/pricing/index.html): Start one assistant with Basic or use Pro for multiple assistant VMs and GPU-powered custom voice or compatible open-weight models such as Gemma. (678 words)
- [First run in the agent workspace | Pneum.ai Docs](/content/docs/agents/first-run/index.html): Complete onboarding, understand the workspace navigation, and verify chat, voice, models, and persistence. (388 words)
- [Launch a Cloud agent | Pneum.ai Docs](/content/docs/cloud/launch-agent/index.html): Deploy a managed Pneum.ai runtime from the Cloud Console and understand every stage from estimate to readiness. (457 words)
- [Schedule agent uptime | Pneum.ai Docs](/content/docs/cloud/schedules/index.html): Configure a timezone-aware weekly window that starts the VM early and stops it automatically. (307 words)
- [Conversation context and memory | Pneum.ai Docs](/content/docs/agents/context-and-memory/index.html): Understand transcript context, automatic compaction, profile separation, notes, and channel-specific conversations. (298 words)
- [Usage, reconciliation, and invoices | Pneum.ai Docs](/content/docs/billing/usage-and-invoices/index.html): Read billing periods, per-agent usage, managed-model totals, invoices, and delayed provider reconciliation. (305 words)
- [Chat and voice conversations | Pneum.ai Docs](/content/docs/agents/chat-and-voice/index.html): Talk or type to your agent, understand activity and replay, and choose push-to-talk or wake mode. (372 words)
- [Network access for Cloud agents | Pneum.ai Docs](/content/docs/cloud/network-access/index.html): Decide between an internet endpoint and Tailscale, manage IP allowlists, and expose only required services. (382 words)
- [Notes Studio | Pneum.ai Docs](/content/docs/studios/notes/index.html): Create, import, organize, search, and reuse Markdown notes and their related media. (287 words)
- [First-agent quickstart | Pneum.ai Docs](/content/docs/getting-started/quickstart/index.html): Launch a recommended CPU agent, wait for readiness, and open the assistant workspace for the first time. (328 words)
- [Start, stop, and terminate agents | Pneum.ai Docs](/content/docs/cloud/lifecycle/index.html): Operate an agent safely, understand retained storage, and know when an action is reversible. (332 words)
- [Plans and agent limits | Pneum.ai Docs](/content/docs/billing/plans/index.html): Compare Basic and Pro behavior, active-agent limits, CPU and GPU access, and separate usage charges. (294 words)
- [Documentation overview | Pneum.ai Docs](/content/docs/getting-started/overview/index.html): Understand the Pneum.ai product, the Cloud Console, and the path from account creation to a working agent. (329 words)
- [Managed models, BYOK, and spending caps | Pneum.ai Docs](/content/docs/billing/managed-models/index.html): Choose how model requests are authorized and billed, then set a monthly limit for managed usage. (288 words)
- [Understand infrastructure estimates | Pneum.ai Docs](/content/docs/billing/cost-estimates/index.html): Read machine, GPU, and disk estimates before launch and distinguish live estimates from reconciled charges. (248 words)
- [Blog | Pneum.ai](/content/blog/index.html): Read Pneum.ai launch notes, design essays, product updates, and voice AI commentary. (79 words)
- [One AI Agent, Three Apps: A Morning Brief From Outlook, Calendar, and Slack | Pneum.ai](/content/blog/ai-agent-morning-brief-outlook-calendar-slack/index.html): See how Pneum.ai can turn Outlook, calendar, and Slack activity into one decision-focused morning brief while keeping you in control of anything it sends or changes. (845 words, 2026-08-17)
- [Launching Pneum.ai | Pneum.ai](/content/blog/launching-pneum-ai/index.html): Pneum.ai is a private realtime voice AI platform with local transcripts, local or hosted LLM support, voice generation, subagent orchestration, and LM Studio compatibility. (617 words, 2026-03-11)

## About Pages

- [About | Pneum.ai](/content/about/index.html): Meet Pneum.ai founder Tyler Grant and co-founder Matthew F., and learn why they are building a more useful everyday assistant. (281 words)

## Resources

- [Full Page Index](/index.html): Browse all cached pages with rich metadata
- [About This Cache](/about.html): Methodology, technical details, and usage guidelines
- [XML Sitemap](/sitemap.xml): Machine-readable sitemap for crawler discovery
- [Robots.txt](/robots.txt): Crawler directives
- [AgentSite Network](https://agentsite.network/network.html): Public index of AgentSites and their machine-readable resources