v1.32.2

This release focuses on security improvements and bug fixes, including prevention of PATH hijacking vulnerabilities and fixes to environment file support.

Bug Fixes

• Fixes prevention of PATH hijacking and TOCTOU (Time-of-Check-Time-of-Use) vulnerabilities in shell/binary resolution (CWE-426)
• Fixes --env-file support for the gateway

Technical Changes

• Removes debug code from codebase
• Reverts user prompt options feature that was previously added

---

What's Changed

• docs: update CHANGELOG.md for v1.32.1 by @docker-read-write[bot] in #2084
• fix: prevent PATH hijacking and TOCTOU in shell/binary resolution by @dgageot in #2083
• Remove debug code by @dgageot in #2086
• Fix --env-file support for the gateway by @dgageot in #2085
• Revert "Add options-based selection to user_prompt tool" by @trungutt in #2088

Full Changelog: v1.32.1...v1.32.2

v1.32.1

This release fixes several issues with session handling, tool elicitation, and MCP environment variable validation.

Bug Fixes

• Fixes corrupted session history by filtering sub-agent streaming events from parent session persistence
• Fixes elicitation requests failing in sessions with ToolsApproved=true by decoupling elicitation channel from ToolsApproved flag
• Fixes MCP environment variable validation being skipped when any gateway preflight errors occur

Improvements

• Prevents sidebar from scrolling to top when clicking navigation links in documentation

Technical Changes

• Adds end-to-end test for tool result block validation
• Updates CHANGELOG.md for v1.32.0 release

---

What's Changed

• docs: update CHANGELOG.md for v1.32.0 by @docker-read-write[bot] in #2072
• Don't scroll sidebar to the top by @dgageot in #2076
• Fix corrupted session history by @dgageot in #2077
• Fix MCP env var check skipped when any gateway preflight errors by @dgageot in #2081
• fix: decouple elicitation channel from ToolsApproved flag by @dgageot in #2080

Full Changelog: v1.32.0...v1.32.1

v1.32.0

This release adds support for newer Gemini models, improves toolset documentation, and enhances user interaction capabilities.

What's New

• Adds options-based selection to user_prompt tool, allowing the agent to present users with labeled choices instead of free-form input
• Documents {ORIGINAL_INSTRUCTIONS} placeholder for enriching toolset instructions rather than replacing them

Bug Fixes

• Fixes support for Gemini 3.x versioned models (e.g., gemini-3.1-pro-preview) to ensure proper model recognition and thinking configuration
• Fixes gateway handling when using docker agent without a command
• Fixes broken links in documentation

Technical Changes

• Adds check for broken links in CI
• Updates .gitignore to exclude cagent-* binaries from being committed

---

What's Changed

• docs: update CHANGELOG.md for v1.31.0 by @docker-read-write[bot] in #2063
• doc: document {ORIGINAL_INSTRUCTIONS} placeholder for toolset instructions by @dgageot in #2062
• fix: support Gemini 3.x versioned models (e.g., gemini-3.1-pro-preview) by @dgageot in #2054
• Fix gateway handling with docker agent without command by @dgageot in #2064
• Fix broken links by @dgageot in #2067
• gitignore cagent-* binaries by @derekmisler in #2069
• Check for broken links by @dgageot in #2068
• Add options-based selection to user_prompt tool by @trungutt in #2071

Full Changelog: v1.31.0...v1.32.0

v1.31.0

This release enhances the cost dialog with detailed session statistics and improves todo tool reliability for better task completion tracking.

What's New

• Adds total token count, session duration, and message count to cost dialog
• Adds reasoning tokens display for supported models (e.g. o1)
• Adds average cost per 1K tokens and per message metrics to cost analysis
• Adds cost percentage breakdown per model and per message
• Adds cache hit rate and per-entry cached token count display

Improvements

• Improves todo tool reliability by reminding LLM of incomplete items and including full state in all responses

Bug Fixes

• Fixes Sonnet model name
• Fixes various edge-case bugs in cost dialog formatting

Technical Changes

• Adds cache to building hub image in CI
• Optimizes CI by building and testing Go on the same runner to avoid duplicate compilation
• Freezes config to v6
• Deduplicates tool documentation into individual pages
• Adds docs-serve task for local Jekyll preview via Docker

---

What's Changed

• docs: update CHANGELOG.md for v1.30.1 by @docker-read-write[bot] in #2050
• Add cache to building hub image in CI by @gtardif in #2037
• cost dialog: enrich with session stats, per-model percentages, and formatting fixes by @dgageot in #2046
• Fix sonnet model name by @dgageot in #2052
• fix: improve todo completion reliability by @trungutt in #2048
• Freeze config v6 by @dgageot in #2059
• Improve the toolsets documentation by @dgageot in #2056

Full Changelog: v1.30.1...v1.31.0

v1.30.1

This release improves command history handling, adds sound notifications, and includes various bug fixes and performance optimizations.

What's New

• Adds sound notifications for long-running tasks and errors (opt-in feature, disabled by default)
• Adds LSP multiplexer to support multiple LSP toolsets simultaneously
• Adds per-toolset model routing via model field on toolsets configuration
• Adds click-to-copy functionality for working directory in TUI sidebar
• Makes background_agents a standalone toolset that can be enabled independently

Improvements

• Improves tmux experience with better keyboard enhancements and focus handling
• Optimizes BM25 scoring strategy for better performance
• Reduces redundant work during evaluation runs
• Fixes animated spinners inside terminal multiplexers
• Repaints terminal on focus to fix broken display after tab switch in Docker Desktop

Bug Fixes

• Fixes loading very long lines in command history that previously caused crashes
• Fixes LSP server being killed by context cancellation and restart failures
• Fixes session-pinned agent usage in RunStream instead of shared currentAgent
• Fixes sidebar context percentage flickering during sub-agent transfers
• Fixes concurrent map writes by moving registerDefaultTools to constructor
• Returns clear error when OPENAI_API_KEY is missing for speech-to-text

Technical Changes

• Splits monolithic runtime.go into focused files by concern
• Refactors code to use slices and maps stdlib functions instead of manual implementations
• Enables modernize and perfsprint linters with all findings resolved
• Migrates tool output to structured JSON schemas for todo tools
• Replaces json.MarshalIndent with json.Marshal in builtin tools
• Uses errors.AsType consistently instead of errors.As with pre-declared variables

---

What's Changed

• Don't ignore GITHUB_TOKEN by @dgageot in #2002
• history: Fix loading very long lines by @vvoland in #1940
• docs: update CHANGELOG.md for v1.30.0 by @docker-read-write[bot] in #2003
• Fix broken links to pages subsections by @gtardif in #2005
• feat: add sound notifications for task completion and errors by @Mostamhd in #1870
• Add LSP multiplexer to support multiple LSP toolsets by @dgageot in #1970
• fix(#2012): Return clear error when OPENAI_API_KEY is missing for speech-to-text by @aheritier in #2013
• Bump direct Go dependencies by @dgageot in #2011
• Fix LSP server killed by context cancellation and restart failures by @dgageot in #2008
• Replace duplicated mockEnvProvider test types with shared environment providers by @dgageot in #2014
• fix: use session-pinned agent in RunStream instead of shared currentAgent by @dgageot in #2009
• docs: fix hallucinated CLI flags, commands, and config formats by @dgageot in #2020
• refactor: use slices and maps stdlib functions instead of manual implementations by @dgageot in #2021
• refactor(anthropic): deduplicate sequencing, media-type, and test helpers by @dgageot in #2019
• refactor: split runtime.go and extract pkg/modelerrors by @dgageot in #2010
• tui: improve tmux experience and simplify keyboard enhancements by @dgageot in #2017
• codemode: fix Start() fail-fast and use tools.As for wrapper unwrapping by @dgageot in #2007
• Simplify rulebased router: remove redundant types and score aggregation by @dgageot in #2016
• Unify streamAdapter/betaStreamAdapter retry logic into generic retryableStream by @dgageot in #2018
• Fix task deploy-local by @dgageot in #2024
• fix: default sound notifications to off (opt-in) by @dgageot in #2025
• feat: add per-toolset model routing via model field on toolsets by @dgageot in #2015
• refactor: use errors.AsType consistently instead of errors.As with pre-declared variables by @dgageot in #2028
• tui: repaint terminal on focus to fix broken display after tab switch by @dgageot in #2026
• Enable modernize and perfsprint linters, fix all findings by @dgageot in #2027
• update Slack link in readme by @derekmisler in #2032
• Replace json.MarshalIndent with json.Marshal in builtin tools by @dgageot in #2031
• refactor(dmr): split client.go into focused files by concern by @dgageot in #2029
• Fix last brew install cagent mention by @gtardif in #2034
• tui: fix animated spinners inside terminal multiplexers by @dgageot in #2035
• refactor(runtime): split monolithic runtime.go into focused files by @dgageot in #2030
• feat: click to copy working directory in TUI sidebar by @dgageot in #2036
• Use ResultSuccess/ResultError helpers in tasks and user_prompt tools by @dgageot in #2040
• fix: move registerDefaultTools to constructor to prevent concurrent map writes by @dgageot in #2041
• Fix sidebar context % flickering during sub-agent transfers by @dgageot in #2042
• eval: reduce redundant work during evaluation runs by @dgageot in #2047
• perf: optimize BM25 scoring strategy by @dgageot in #2043
• refactor: remove duplication in model resolution, thinking budget, and message construction by @dgageot in #2038
• feat: make background_agents a standalone toolset by @dgageot in #2033
• todo: migrate tool output to structured JSON schemas by @dgageot in #2045

Full Changelog: v1.30.0...v1.30.1

v1.30.0

This release introduces file drag-and-drop support, background agent tasks, and completes the transition from "cagent" to "docker-agent" branding throughout the codebase.

What's New

• Adds file drag-and-drop support for images and PDFs with visual file type indicators and 5MB size limit per file
• Adds background agent task tools (run_background_agent, list_background_agents, view_background_agent, stop_background_agent) for concurrent sub-agent dispatch
• Adds --sandbox flag to run command for Docker sandbox isolation
• Adds model_picker toolset for dynamic model switching between LLM models mid-conversation
• Adds search, update, categories, and default path functionality to memory tool
• Adds MiniMax as a built-in provider alias with MINIMAX_API_KEY support
• Adds top-level mcps section for reusable MCP server definitions in agent configs
• Adds support for OCI/catalog and URL references as sub-agents and handoffs

Improvements

• Auto-continues max iterations in --yolo mode instead of prompting
• Improves toolset error reporting to show specific toolset information
• Improves user_prompt TUI dialog with title, free-form input, and navigation
• Auto-pulls DMR models in non-interactive mode
• Animates window title while working for tmux activity detection
• Supports comma-separated string format for allowed-tools in skills

Bug Fixes

• Fixes thread blocking when attachment file is deleted
• Fixes max iterations handling in JSON output mode
• Fixes text to speech on macOS
• Fixes context window overflow with auto-recovery and proactive compaction
• Fixes data races in Session Messages slice and test functions
• Fixes SSE streaming by disabling automatic gzip compression
• Applies ModifiedInput from pre-tool hooks to tool call arguments

Technical Changes

• Completes rename from "cagent" to "docker-agent" throughout codebase, documentation, and repository URLs
• Supports both DOCKER_AGENT_* and legacy CAGENT_* environment variables
• Removes --exit-on-stdin-eof flag and ConnectRPC code
• Adds timeouts to shutdown contexts to prevent goroutine leaks
• Extracts TodoStorage interface with in-memory implementation
• Refactors listener lifecycle to return cleanup functions
• Updates Dockerfile to use docker-agent binary with cagent as compatible symlink

---

What's Changed

• docs: update CHANGELOG.md for v1.29.0 by @docker-read-write[bot] in #1909
• Bump Go dependencies by @dgageot in #1913
• update taskfile to build both images docker/cagent and docker/docker-agent by @gtardif in #1916
• Remove --exit-on-stdin-eof by @dgageot in #1918
• Rename env vars CAGENT_ to DOCKER_AGENT_ (keep support for old env vars) by @gtardif in #1917
• agent: Improve toolset error reporting by @vvoland in #1914
• Nightly scanner should be less nit-picky about docs by @derekmisler in #1921
• Update docs and samples to rename docker-agent, change usage samples to docker agent by @gtardif in #1915
• Simplify the AGENTS.md a LOT by @dgageot in #1923
• Fix a few issues in the docs by @dgageot in #1924
• docs: remove generated pages/ from git tracking by @dgageot in #1927
• Fix speech to text on macOS by @dgageot in #1922
• More docs rename (in / docs), fix remaining docker agent serve a2a/acp/mcp by @gtardif in #1928
• Fix test by @dgageot in #1929
• Fix #1911 by @dgageot in #1931
• Validate yaml in doc by @dgageot in #1932
• Fix a few race conditions seen in tests by @dgageot in #1930
• Fix broken links in docs pages, were not using relative urls by @gtardif in #1939
• Rename CAGENT_HIDE_TELEMETRY & CAGENT_EXP_DEBUG_LAYOUT. Still support old env vars by @gtardif in #1926
• Teach the agent to work with our config versions by @dgageot in #1938
• Improve pkg/js by @dgageot in #1933
• Improve README by @dgageot in #1936
• Add model_picker toolset for dynamic model switching by @dgageot in #1937
• cagent run --sandbox by @dgageot in #1904
• Support auto-downloading tools by @dgageot in #1925
• Leverage latest MCP spec features from go-sdk v1.4.0 by @dgageot in #1955
• bump google.golang.org/adk from v0.4.0 to v0.5.0 by @dgageot in #1954
• Animate window title while working for tmux activity detection by @dgageot in #1949
• Fix listener resource leaks in serve commands by @dgageot in #1945
• Show the new TUI by @dgageot in #1942
• Auto-pull DMR models in non-interactive mode by @dgageot in #1944
• Improve user_prompt TUI dialog: title, free-form input, and navigation by @dgageot in #1943
• Improve sub-sessions usage by @dgageot in #1941
• Add MiniMax as a built-in provider alias by @dgageot in #1948
• Bump go dependencies by @dgageot in #1953
• Rename repo URL and pages URL by @gtardif in #1957
• Support OCI/catalog and URL references as sub-agents and handoffs by @dgageot in #1946
• Improve docs search by @dgageot in #1959
• Use docker agent command by @dgageot in #1958
• todo: extract storage interface with in-memory implementation by @rumpl in #1960
• docker-agent is primary binary in taskfile by @gtardif in #1961
• Add top-level mcps section for reusable MCP server definitions by @dgageot in #1947
• A few more renames from cagent by @dgageot in #1962
• Some more cagent urls by @gtardif in #1964
• fix(cli): auto-continue max iterations in --yolo mode by @frntn in #1737
• fix(editor): prevent thread block when attachment file is deleted by @Pnkcaht in #1736
• Fix main branch by @dgageot in #1968
• feat: add file drag-and-drop support for images and PDFs by @ambigois in #1658
• Add background agent task tools for concurrent sub-agent dispatch (#863) by @jfrey in #1908
• fix(hooks): apply ModifiedInput from pre-tool hooks to tool call arguments by @aheritier in #1950
• Add timeouts to shutdown contexts to prevent goroutine leaks by @dgageot in #1965
• Update winget workflow to modify Docker.Agent package, with the new GH repo name by @gtardif in #1972
• updated GHA with new checks:write permission by @derekmisler in #1974
• Disable automatic gzip compression to fix SSE streaming by @dgageot in #1967
• Fix cobra command and rename more things from cagent to docker agent by @dgageot in #1979
• Support comma-separated string for allowed-tools in skills by @dgageot in #1984
• Fix documentation by @dgageot in #1983
• Fix model name by @dgageot in #1991
• Fix gopls versions by @dgageot in #1988
• Daily fixes by @dgageot in #1990
• auto-complete tests by @gtardif in #1989
• Dockerfile with docker-agent binary, keeping cagent only as compatible symlink by @gtardif in #1992
• Rename cagent in eval by @gtardif in #1993
• Fix documentation by @dgageot in #1995
• More renames from cagent to docker-agent by @dgageot in #1994
• Rename e2e test files by @dgageot in #1997
• Remove useless documentation by @dgageot in #1998
• Remove ConnectRPC code by @dgageot in #1996
• Add search, update, categories, and default path to memory tool by @nunocoracao in #1971
• More renames by @dgageot in #1999
• Fix context window overflow: auto-recovery and proactive compacti...

v1.29.0

This release adds automated issue triage capabilities and new CLI configuration options for directory overrides.

What's New

• Adds auto issue triage workflow that automatically evaluates bug reports and can create fix PRs
• Adds --config-dir, --data-dir, and --cache-dir global CLI flags to override default paths

Bug Fixes

• Fixes result marker parsing in auto-issue-triage workflow to handle LLM output with trailing empty lines
• Fixes GitHub Pages deployment issues

Technical Changes

• Updates nightly scanner documentation and configuration
• Removes draft status from PR creation workflow steps
• Adds tip about the default agent in documentation

---

What's Changed

• docs: update CHANGELOG.md for v1.28.1 by @docker-read-write[bot] in #1902
• Fix the github pages? by @dgageot in #1903
• Fix GitHub pages deployment by @dgageot in #1901
• feat: add auto issue triage workflow by @derekmisler in #1888
• Replace the brittle tail -n 1 parsing with something that searches for the marker by @derekmisler in #1905
• Add tip about the default agent by @dgageot in #1906
• Add --config-dir and --data-dir global CLI flags to override default paths by @dgageot in #1907

Full Changelog: v1.28.1...v1.29.0

v1.28.1

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

v1.28.0

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

v1.27.1

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