# Agent Zero Release Notes ## Highlights This release introduces a redesigned first-run onboarding experience, separates Desktop and Office into focused plugins, adds host browser routing through the A0 CLI connector, and delivers significant prompt and tool contract improvements informed by cross-provider evidence sweeps. --- ## New Features - **Guided first-run onboarding** — A new wizard walks users through Cloud vs. Local provider selection, account connection, model picking, and a ready state, replacing the previous manual setup flow. - **Ollama Cloud provider** — Available as an OpenAI-compatible chat provider with hosted API base and model listing. - **Nebius Token Factory provider** — Registered as an OpenAI-compatible provider, available in onboarding and model config. - **Host Browser (Bring Your Own Browser)** — Route browser actions through the A0 CLI connector to a local Chrome/Chromium instance with remote-debugging support, including profile mode selection (existing profile or clean Agent profile). - **Browser runtime selection exposed to CLI** — A protected connector endpoint allows the A0 CLI to switch between Docker browser and Bring Your Own Browser mode. - **Persistent browser history screenshots** — Each Browser tool call saves a static JPEG to chat history for immutable transcript previews. - **ZIP download toast feedback** — Immediate grouped progress toasts for backup and file-browser ZIP downloads with error replacement. - **API chat lifetime and cleanup** — `lifetime_hours` is validated and persisted per chat; a background job loop removes expired API chats automatically. - **Connector model overrides in Web UI** — Custom per-chat model overrides now render in the model switcher instead of a generic "Custom" label, with live refresh on changes. - **`chats-header-controls` extension hook** — Plugins can inject controls into the chats list header via the established `x-extension` pattern. ## Architecture & Plugin Changes - **Desktop split into `_desktop` plugin** — Xpra/Xfce lifecycle, `/desktop` route, Desktop state, session APIs, surface registration, and the linux-desktop skill now live in `_desktop`. - **Office reduced to document ownership** — `_office` focuses on document artifacts, Markdown sessions, LibreOffice-compatible file actions, and persistence. Binary document editing routes through explicit Desktop requests. - **Shared surfaces frontend service** — Browser and Desktop register docked or floating live UI through a dedicated surfaces layer, preserving active viewers across canvas/modal switches. - **Office and Desktop state moved under plugin storage** — Retired `/usr/_office` and `/usr/_desktop` trees migrate into `/usr/plugins/<plugin>` at startup. - **Serialized runtime package preparation** — Office and Desktop apt operations share an in-process retry guard to handle transient apt locks gracefully. - **Desktop install progress** — The Desktop frontend shows a neutral install message during runtime preparation instead of a misleading missing-dependencies error. ## Tool & Prompt Improvements - **Normalized tool action contracts** — Multi-action tools standardized around `tool_args.action` while maintaining backward compatibility with older dialects. - **Renamed skills to task-oriented names** — Office skills become Writer, Calc, Impress; high-impact skills get task-oriented names; plugin-owned skills move into their owning plugin folders. - **Reduced default prompt pressure** — Browser and scheduler prompts compacted into skill-backed manifests, skill catalog descriptions shortened, noisy framework knowledge pruned. - **A0 connector remote skills split by affordance** — Separate text-editor and code-execution remote skills replace the combined CLI remote workflow skill; computer-use remote scoped to desktop control and skill-gated as beta. - **Browser tool ergonomics** — Selector-based reference actions, coordinate click fallbacks, focused-field typing, string key chords, and smarter content helper snapshots that ignore framework event wrappers. - **Document artifact stabilization** — File creation is opt-in through `document_artifact`, file cards appear only in final responses, and Desktop canvas sessions refresh after saved edits. ## Security Fixes - **Path-containment bypass in `/api/image_get` resolved** — Requested images are now resolved against the Agent Zero base directory with symlink-aware validation. SVG/SVGZ responses include `nosniff` and sandboxed CSP headers. ## Bug Fixes - **Blocking history compression edge cases** — Stalled automatic compression no longer spins forever; large single-line histories (85k+ characters) split by verified token budget. - **Canvas attachment for browser and documents** — Browser canvas attaches to active Docker sessions with initial snapshots; Markdown documents open through the right-canvas Desktop editor. - **Bare desktop canvas treated as ready** — Linux Desktop reports healthy when XFCE has no active application window but display and screenshots are available. - **Computer-use rearm handling** — Stale Free Run settings no longer treated as ready; agents provide backend-agnostic rearm guidance. - **Subordinate agent document affordances** — Automatic office document response affordances skipped for delegated agents so reviews keep their actual content. - **Model config provider controls** — Custom API base URL resets on provider change; Supports Vision toggle moved out of Advanced Settings. ## Infrastructure & Maintenance - **Browser Playwright cache moved to `/a0/tmp/playwright`** — Startup migration cleans retired usr-based caches automatically. - **uv cache cleaned during self-update** — Best-effort `uv cache clean` runs when the self-update manager processes an update. - **Legacy Collabora retirement** — Old `a0_office_collabora` supervisor program stopped and removed during Docker self-update; stale runtime paths cleaned. - **Legacy runtime artifact cleanup** — Stale directories removed, retired packages (including dynamically discovered `collaboraoffice*` splits) purged at startup/self-update. - **Self-update backups skip runtime sockets** — Live sockets, device nodes, vanished files, and unreadable entries no longer trip rollback checks. - **Connector hello metadata** — Remote tool metadata stored through a dedicated hello path with context-bound websocket SID acknowledgment. - **Skill ownership hardened** — Connector list/delete scoped to enabled catalog; built-in plugin skills cannot be deleted; invalid skill YAML surfaced instead of silently accepted. ## UI Improvements - **Simplified project LLM setup** — Project creation focuses on basic details and Git cloning; model selection moved to a collapsed Advanced Settings accordion on the edit screen. - **Settings accordion redesign** — Sticky accordion with active section tracking replaces two-step settings navigation. - **Action button order aligned** — Primary actions appear before Cancel across plugin settings and related modals. - **Browser chat action button removed** — The Browser plugin no longer injects a button under the chat input. - **Project skills settings repositioned** — Moved directly after project instructions, before file structure and secrets. - **Host Browser config simplified** — "Use host when ready" replaced with clear "Bring Your Own Browser" option and Chrome remote-debugging guidance. ## Documentation - **Refreshed user-facing guides** — README, quickstart, and docs index updated around Browser, Desktop, A0 CLI, projects, memory, skills, profiles, and model presets with optimized screenshots. - **First-run onboarding guide** — Screenshot-led walkthrough linked from docs hub, quickstart, and installation flow. - **Model efficiency scorecard** — Scored tool-efficiency chart with provider/model rankings, failure clusters, and applied improvements. - **Refactor guardrails** — Internal docs covering modal/surface boundary, Desktop ownership, Office scope, renamed skills, and connector ownership rules.

## Release Notes - **Fixed PTY file descriptor leaks in code execution** — POSIX PTY master descriptors are now properly closed when terminal sessions end, preventing `/dev/ptmx` descriptor exhaustion under sustained use. Additionally, closed or exited PTY sessions are detected before writes, with automatic retry/recovery on failure, and double-close of reused file descriptors is prevented through idempotent shutdown handling. - **Browser runtime recovers from stale Playwright contexts** — Cached browser contexts that have already closed are now detected before reuse. Stale pages, screencasts, and interaction state are cleared, and the Playwright instance restarts cleanly on next use. - **Desktop URLs open in the Agent Zero Browser** — Clicking URLs in the Xfce Desktop environment now routes them into the Browser tool on the opposite canvas/modal surface. Floating Browser and Desktop modals also pass through outside clicks while preserving interaction inside the modal window. - **Self-update no longer fails on broken symlinks** — Dangling symlinks in `/usr` (e.g., host-created virtualenv links) are logged and skipped during backup, allowing updates to complete without aborting. - **Deferred Office/Desktop startup** — The Desktop environment now starts only when the Desktop surface is opened or an Office document is created/opened, rather than eagerly on mount. A loading indicator provides immediate feedback while Xpra initializes. - **ARM64 Xpra codec gaps handled gracefully** — Optional local Xpra GUI client packages are treated as best-effort on ARM64, so missing codec dependencies no longer surface as startup warnings when the browser-hosted Desktop is already functional.

### v0.9.8.2 - Bugfix - skills loading ### v0.9.8.1 - Caching, context optimization, error handling... - improved caching for anthropic models - improved context window optimimization - fixed retry on critical error counter - load up to 5 skills at a time - default chat model is now Sonnet 4.6 ### v0.9.8 - Skills, UI Redesign & Git projects [Release video](https://youtu.be/NV7s78yn6DY) - Skills - Skills System replacing the legacy Instruments with a new `SKILL.md` standard for structured, portable agent capabilities. - Built-in skills, and UI support for importing and listing skills - Real-time WebSocket infrastructure replacing the polling-based approach for UI state synchronization - UI Redesign - Process groups to visually group agent actions with expand/collapse support - Timestamps, steps count and execution time with tool-specific badges - Step detail modals with key-value and raw JSON display - Collapsible responses with show more/less and copy buttons on code blocks and tables - Message queue system allowing users to queue messages while the agent is still processing - In-browser file editor for viewing and editing files without leaving the UI - Welcome screen redesign with info and warning banners for connection security, missing API keys, and system resources - Scheduler redesign with standalone modal, separate task list, detail and editor components, and project support - Smooth response rendering and scroll stabilization across chat, terminals, and image viewer - Chat width setting and reworked preferences panel - Image viewer improvements with scroll support and expanded viewer - Redesigned sidebar with reusable dropdown component and streamlined buttons - Inline button confirmations for critical actions - Improved login design and new logout button - File browser enhanced with rename and file actions dropdown - Git projects - Git-based projects with clone authentication for public and private repositories - Four new LLM providers: CometAPI, Z.AI, Moonshot AI, and AWS Bedrock - Microsoft Dev Tunnels integration for secure remote access - User data migration to `/usr` directory for cleaner separation of user and system files - Subagents system with configurable agent profiles for different roles - Memory operations offloaded to deferred tasks for better performance - Environment variables can now configure settings via `A0_SET_*` prefix in `.env` - Automatic migration with overwrite support for `.env`, scheduler, knowledge, and legacy directories - Projects support extended to MCP, A2A, and external API - Workdir outside project support for more flexible file organization - Agent number tracking in backend and responses for multi-agent identification - Many bug fixes and stability improvements across the UI, MCP tools, scheduler, uploads, and WebSocket handling

### v0.9.8.1 - Caching, context optimization, error handling... - improved caching for anthropic models - improved context window optimimization - fixed retry on critical error counter - load up to 5 skills at a time - default chat model is now Sonnet 4.6 ### v0.9.8 - Skills, UI Redesign & Git projects [Release video](https://youtu.be/NV7s78yn6DY) - Skills - Skills System replacing the legacy Instruments with a new `SKILL.md` standard for structured, portable agent capabilities. - Built-in skills, and UI support for importing and listing skills - Real-time WebSocket infrastructure replacing the polling-based approach for UI state synchronization - UI Redesign - Process groups to visually group agent actions with expand/collapse support - Timestamps, steps count and execution time with tool-specific badges - Step detail modals with key-value and raw JSON display - Collapsible responses with show more/less and copy buttons on code blocks and tables - Message queue system allowing users to queue messages while the agent is still processing - In-browser file editor for viewing and editing files without leaving the UI - Welcome screen redesign with info and warning banners for connection security, missing API keys, and system resources - Scheduler redesign with standalone modal, separate task list, detail and editor components, and project support - Smooth response rendering and scroll stabilization across chat, terminals, and image viewer - Chat width setting and reworked preferences panel - Image viewer improvements with scroll support and expanded viewer - Redesigned sidebar with reusable dropdown component and streamlined buttons - Inline button confirmations for critical actions - Improved login design and new logout button - File browser enhanced with rename and file actions dropdown - Git projects - Git-based projects with clone authentication for public and private repositories - Four new LLM providers: CometAPI, Z.AI, Moonshot AI, and AWS Bedrock - Microsoft Dev Tunnels integration for secure remote access - User data migration to `/usr` directory for cleaner separation of user and system files - Subagents system with configurable agent profiles for different roles - Memory operations offloaded to deferred tasks for better performance - Environment variables can now configure settings via `A0_SET_*` prefix in `.env` - Automatic migration with overwrite support for `.env`, scheduler, knowledge, and legacy directories - Projects support extended to MCP, A2A, and external API - Workdir outside project support for more flexible file organization - Agent number tracking in backend and responses for multi-agent identification - Many bug fixes and stability improvements across the UI, MCP tools, scheduler, uploads, and WebSocket handling

# v0.9.6 - Memory Dashboard [Release video](https://youtu.be/sizjAq2-d9s) - Memory Management Dashboard - Kali update - Python update + dual installation - Browser Use update - New login screen - LiteLLM retry on temporary errors - Github Copilot provider support

## 0.9.5.1 - added support for Agent Zero Venice provider - added support for xAI provider ## 0.9.5 - Secrets management - agent can use credentials without seeing them - Agent can copy paste messages and files without rewriting them - LiteLLM global configuration field - Custom HTTP headers field for browser agent - Progressive web app support - Extra model params support for JSON - Short IDs for files and memories to prevent LLM errors - Tunnel component frontend rework - Fix for timezone change bug - Notifications z-index fix

- Secrets management - agent can use credentials without seeing them - Agent can copy paste messages and files without rewriting them - LiteLLM global configuration field - Custom HTTP headers field for browser agent - Progressive web app support - Extra model params support for JSON - Short IDs for files and memories to prevent LLM errors - Tunnel component frontend rework - Fix for timezone change bug - Notifications z-index fix

# v0.9.2 - Kokoro TTS, Attachments - Kokoro text-to-speech integration - New message attachments system - Minor updates: log truncation, hyperlink targets, component examples, api cleanup