This release adds image support for AI agents, improves cross-platform compatibility, and includes various stability fixes.

What's New

• Adds image support to read_file tool and MCP tool results, allowing agents to view and describe images
• Adds content-based MIME detection and automatic image resizing for vision capabilities
• Strips image content for text-only models using model capabilities detection

Improvements

• Reduces builtin tool prompt lengths while preserving key examples for better performance
• Skips hidden directories in recursive skill loading to avoid walking large trees like .git and .node_modules
• Only uses insecure TLS for localhost OTLP endpoints for better security

Bug Fixes

• Fixes Esc key not interrupting sub-agents in multi-agent sessions
• Fixes slice bounds out of range panic for short JWT tokens
• Fixes goroutine tight loop in LSP readNotifications
• Fixes race condition with elicitation events channel
• Avoids looping forever on symlinks during skill loading
• Handles json.Marshal errors for tool Parameters and OutputSchema

Technical Changes

• Replaces syscall.Rmdir with golang.org/x/sys for cross-platform directory removal
• Removes per-chunk UpdateMessage debug log from SQLite store to reduce log noise
• Stops tool sets for team loaded in GetAgentToolCount
• Migrates GitHub pages to markdown with Jekyll

What's Changed

• docs: update CHANGELOG.md for v1.28.0 by @docker-read-write[bot] in #1892
• Replace syscall.Rmdir with golang.org/x/sys for cross-platform directory removal by @rumpl in #1885
• Remove per-chunk UpdateMessage debug log by @trungutt in #1896
• Fixes to the documentation by @dgageot in #1893
• Pushes docker/docker-agent next to docker/cagent hub image by @gtardif in #1897
• Reduce builtin tool prompt lengths while preserving key examples by @dgageot in #1879
• Daily fixes of the bot-detected issues by @dgageot in #1895
• Skip hidden directories in recursive skill loading by @dgageot in #1875
• 👀 Vision 👀 by @rumpl in #1889
• Migrate our GitHub pages to markdown, with Jekyll by @dgageot in #1900
• fix: Esc key not interrupting sub-agents in multi-agent sessions by @dgageot in #1899

Full Changelog: v1.28.0...v1.28.1

This release improves authentication debugging, session management, and MCP server reliability, along with UI enhancements to the command palette.

What's New

• Adds 'debug auth' command to inspect Docker Desktop JWT with optional JSON output
• Adds automatic retry functionality for all models, including those without fallbacks

Improvements

• Improves MCP server lifecycle with caching and auto-restart capabilities using exponential backoff
• Sorts command palette actions alphabetically within each group
• Uses tea.View.ProgressBar instead of raw escape codes for better display

Bug Fixes

• Fixes session derailment by preserving user messages during conversation trimming
• Fixes duplicate Session header in command palette on macOS
• Fixes mcp/notion not working with OpenAI models by properly walking additionalProperties in schemas
• Defaults to string type for script tool arguments when type is not specified

Technical Changes

• Updates tool filtering documentation
• Updates CHANGELOG.md for v1.27.1
• Updates Charm libraries to stable v2.0.0 releases (bubbletea, bubbles, lipgloss)

What's Changed

• docs: update CHANGELOG.md for v1.27.1 by @docker-read-write[bot] in #1862
• Fix mcp/notion not working with OpenAI models by @vvoland in #1871
• Fix script args with DMR by @dgageot in #1859
• Add 'debug auth' command to inspect Docker Desktop JWT by @dgageot in #1861
• Improve MCP server lifecycle: caching and auto-restart by @dgageot in #1872
• fix(#1863): preserve user messages in trimMessages to prevent session derailment by @aheritier in #1864
• Improve tool filtering doc by @dgageot in #1874
• Improve Commands dialog by @dgageot in #1877
• Bump dependencies by @dgageot in #1876
• Add retries even for models without fallbacks by @rumpl in #1886

Full Changelog: v1.27.1...v1.28.0

This release improves the user interface experience with better message editing capabilities and fixes several issues with token usage tracking and session loading.

What's New

• Adds on_user_input hook that triggers when the agent is waiting for user input or tool confirmation

Improvements

• Improves multi-line editing of past user messages
• Adds clipboard paste support during inline message editing
• Makes loading past sessions faster
• Updates TUI display when the current agent changes

Bug Fixes

• Fixes token usage being recorded multiple times per stream, preventing inflated telemetry counts
• Fixes empty inline edit textarea expanding to full height
• Fixes docker ai shellout to cagent for standalone invocations

Technical Changes

• Updates schema tests to only run for latest version
• Fixes documentation issues

What's Changed

• docs: update CHANGELOG.md for v1.27.0 by @docker-read-write[bot] in #1846
• Repaint the TUI when the current agent changes by @dgageot in #1845
• fix: record token usage once per stream to prevent inflated telemetry by @dgageot in #1855
• Make loading past sessions faster by @dgageot in #1854
• Improve editing past user messages by @dgageot in #1850
• Schema tests should be only for latest version by @dgageot in #1857
• feat(hooks): add on_user_input by @doringeman in #1847
• Fix doc by @dgageot in #1858
• Fix docker ai shellout to cagent by @gtardif in #1860

New Contributors

• @doringeman made their first contribution in #1847

Full Changelog: v1.27.0...v1.27.1

This release introduces dynamic agent color styling for multi-agent teams, adds new filesystem tools, and includes several bug fixes and security improvements.

What's New

• Adds dynamic agent color styling system that assigns unique, deterministic colors to each agent in multi-agent teams for visual distinction across the TUI
• Adds hue-based agent color generation with theme integration that adapts saturation and lightness based on theme background
• Adds mkdir and rmdir filesystem tools so agents can create and remove directories without using shell commands
• Allows .github and .gitlab directories in WalkFiles traversal for better CI workflow support

Bug Fixes

• Fixes race condition in agent color style lookups
• Fixes path traversal vulnerability in ACP filesystem operations
• Fixes YAML marshalling issues that could produce corrupted configuration files
• Handles case-insensitive filesystems properly
• Logs errors when persisting session title in TUI

Technical Changes

• Consolidates color utilities into styles/colorutil.go
• Unexports internal color helpers and deduplicates fallbacks
• Fixes cassettes functionality

What's Changed

• docs: update CHANGELOG.md for v1.26.0 by @docker-read-write[bot] in #1841
• Daily fixes by @dgageot in #1832
• tools/fs: Add mkdir and rmdir by @vvoland in #1781
• feat(#1756): Add dynamic agent color styling system by @aheritier in #1757
• allow .github and .gitlab directories in WalkFiles traversal by @maxcleme in #1833
• Fix yaml marshalling by @dgageot in #1844

Full Changelog: v1.26.0...v1.27.0

This is a maintenance release with dependency updates and internal improvements.

Technical Changes

• Maintenance release with dependency updates

Full Changelog: v1.25.0...v1.26.0

This release introduces remote skills discovery capabilities and improves file reading tools with pagination support.

What's New

• Adds remote skills discovery with disk cache and dedicated tools, supporting the well-known skills discovery specification
• Adds offset and line_count pagination parameters to read_file and read_multiple_files tools for incremental reading of large files

Improvements

• Limits output size for read_file and read_multiple_files tools to prevent excessive token usage
• Removes pagination instructions from tool descriptions for cleaner interface

Bug Fixes

• Fixes LineCount metadata on truncated read_multiple_files results

Technical Changes

• Freezes configuration version v5 and bumps to v6
• Updates test cassettes to match schema changes for file reading tools

What's Changed

• docs: update CHANGELOG.md for v1.23.6 by @docker-read-write[bot] in #1829
• builtin: add offset and line_count pagination to read_file and read_multiple_files by @trungutt in #1828
• Implement remote skills discovery with disk cache and dedicated tools by @rumpl in #1822
• Freeze v5 (and a few refactoring) by @dgageot in #1810

Full Changelog: v1.23.6...v1.24.0

This release improves cost tracking accuracy, enhances session management, and fixes several UI and functionality issues.

What's New

• Adds tab completion for /commands dialog
• Adds mouse support for selecting and opening sessions in the sessions dialog

Improvements

• Computes session cost from messages instead of accumulating on session for better accuracy
• Includes compaction cost in /cost dialog
• Displays original YAML model names in sidebar instead of resolved aliases
• Improves emoji copying support by reversing clipboard copy order (OSC52 first, then pbcopy fallback)

Bug Fixes

• Fixes token usage percentage display during and after agent transfers
• Fixes session forking and costs calculation
• Fixes actual provider display for alloy models in sidebar (was showing wrong provider)
• Restores ctrl-1, ctrl-2... shortcuts for quick agent selection
• Fixes NewHandler panic on parameterless tool calls

Technical Changes

• Consolidates TokenUsage event constructors
• Removes dead UpdateLastAssistantMessageUsage method
• Emits TokenUsageEvent on session restore for context percentage display
• Emits TokenUsageEvent after compaction so sidebar cost updates
• Adds e2e tests on binaries for CLI plugin execution
• Creates ~/.docker/cli-plugins directory if it doesn't exist

What's Changed

• docs: update CHANGELOG.md for v1.23.5 by @docker-read-write[bot] in #1803
• Better support copying emojis by @dgageot in #1804
• Fix multiple cost/tokens related issues by @dgageot in #1795
• Tab completion for /commands dialog by @dgageot in #1806
• Update winget workflow by @lorenrh in #1808
• feat: use docker read write bot by @lorenrh in #1813
• Binary e2e tests by @gtardif in #1812
• fix: use actual provider for alloy models in sidebar by @Aliexe-code in #1807
• Improve sessions dialog by @dgageot in #1811
• Bump Go dependencies by @dgageot in #1817
• fix: restore ctrl-1, ctrl-2... shortcuts for quick agent selection by @dgageot in #1816
• Refactor winget workflow to use wingetcreate CLI by @gtardif in #1826
• get_memories errors on new memories by @derekmisler in #1827

New Contributors

• @Aliexe-code made their first contribution in #1807

Full Changelog: v1.23.5...v1.23.6

This release improves the session browser interface and fixes several issues with the docker-agent standalone binary.

Improvements

• Shows message count in session browser dialog for better session overview

Bug Fixes

• Fixes recognition of cobra internal completion commands as subcommands
• Fixes help text display for docker-agent standalone binary exec
• Fixes version output for docker-agent CLI plugin and standalone exec

Technical Changes

• Renames internal schema structure

What's Changed

• docs: update CHANGELOG.md for v1.23.4 by @docker-read-write[bot] in #1792
• Rename schema by @dgageot in #1793
• Show message count in session browser dialog by @dgageot in #1791
• Fix bash/zsh/... completion by @dgageot in #1794
• Fix help for docker-agent standalone binary exec by @gtardif in #1796
• Fix docker-agent version for cli plugin & standalone exec by @gtardif in #1802

Full Changelog: v1.23.4...v1.23.5

This release introduces parallel session support with tab management, major command restructuring, and enhanced UI interactions.

What's New

• Adds parallel session support with a new tab view to switch between sessions
• Adds drag and drop functionality for reordering tabs
• Adds mouse click support to elicitation, prompt input, and tool confirmation dialogs
• Adds X-Cagent-Model-Name header to models gateway requests
• Adds Ask list to permissions config to force confirmation for read-only tools
• Defaults to running the default agent when no subcommand is given

Improvements

• Restores ctrl-r binding for searching prompt history
• Updates Claude Sonnet model version to 4.6
• Prevents closing the last remaining tab with Ctrl+W
• Makes fetch tool not read-only
• Handles Claude overloaded_error with retry logic

Bug Fixes

• Fixes ctrl-c in docker agent and docker agent defaulting to docker agent run
• Fixes completion command
• Fixes cagent-action to expect a prompt
• Fixes gemini use of vertexai environment variables
• Fixes CPU profile file handling and error handling in isFirstRun

Technical Changes

• Removes cagent config commands (breaking change)
• Removes cagent feedback command (breaking change)
• Removes cagent build command (breaking change)
• Removes cagent catalog command (breaking change)
• Moves a2a, acp, mcp and api commands under cagent serve (breaking change)
• Replaces cagent exec with cagent run --exec (breaking change)
• Moves pull and push under cagent share (breaking change)
• Hides cagent debug
• Adds skills to the default agent
• Defaults restore_tabs to false

What's Changed

• docs: update CHANGELOG.md for v1.23.3 by @docker-read-write[bot] in #1753
• Parallel sessions! Tabs! by @krissetto in #1695
• Restore ctrl-r binding for searching prompt history by @krissetto in #1759
• Review cagent commands by @dgageot in #1755
• default restore_tabs to false by @krissetto in #1760
• Fix completion command by @dgageot in #1761
• fix: cagent-action expects a prompt by @derekmisler in #1762
• feat: add X-Cagent-Model-Name header to models gateway requests by @tdabasinskas in #1751
• chore(config): Update Claude Sonnet model version to 4.6 by @aheritier in #1768
• Add mouse click support to elicitation, prompt input, and tool confirmation dialogs by @dgageot in #1766
• temp home dir to avoid issues in some environments by @krissetto in #1773
• drag 'n drop tabs by @krissetto in #1772
• fix: gemini use of vertexai environment variables by @abaez-pantheon in #1763
• fallback: Handle overloaded_error by @vvoland in #1780
• permissions: add Ask list to force confirmation for tools by @dgageot in #1785
• Bump Go dependencies by @dgageot in #1777
• Fix ctrl-c in docker agent serve api and fix docker agent defaulting to docker agent run by @dgageot in #1782
• Make fetch tool not read-only by @dgageot in #1786
• Keep same error handling for main cli plugin execution by @gtardif in #1789
• tui/tabbar: Prevent closing the last remaining tab by @vvoland in #1790
• Fix path and typo by @dgageot in #1788
• Daily fixes for the Nightly issue detector by @dgageot in #1787

New Contributors

• @abaez-pantheon made their first contribution in #1763

Full Changelog: v1.23.3...v1.23.4

This release adds Docker CLI plugin support and improves TUI performance by making model reasoning checks asynchronous.

What's New

• Adds support for using cagent as a Docker CLI plugin with docker agent command (no functional changes to existing cagent command)
• Handles Windows .exe binary suffix for CLI plugin compatibility

Improvements

• Makes model reasoning support checks asynchronous to prevent TUI freezing (previously could block for up to 30 seconds)
• Threads context.Context through modelsdev store API to allow proper cancellation and deadline propagation

Technical Changes

• Renames cagent OCI annotation to io.docker.agent.version while maintaining backward compatibility with the old annotation
• Updates config media type to use docker.agent
• Adds TUI general guidelines to AGENTS.md documentation

What's Changed

• Thread context.Context through modelsdev store API by @rumpl in #1747
• Rename cagent OCI annotation, keep old one by @gtardif in #1745
• docs: update CHANGELOG.md for v1.23.2 by @docker-read-write[bot] in #1746
• Move ModelSupportsReasoning calls to async bubbletea commands by @rumpl in #1749
• Allow to use cagent binary as a docker cli plugin docker-agent. No functional change for cagent command. by @gtardif in #1748

New Contributors

• @gtardif made their first contribution in #1745

Full Changelog: v1.23.2...v1.23.3