task-master-ai@0.41.0

useautumn/autumntask-master-ai@0.41.0Jan 11, 2026by github-actions[bot]

AI Summary

This release introduces the new `loop` command for automated task execution with Claude Code, featuring built-in presets and progress tracking, alongside fixes for OpenAI JSON schema compatibility.

Key Highlights

  • Added `task-master loop` command for automated execution
  • Added tool annotations (readOnlyHint, destructiveHint) for LLMs
  • Fixed OpenAI structured output validation by removing `.default()` from Zod schemas
  • Added progress file tracking

New Features

  • `loop` command with Docker sandbox support
  • Presets for `default`, `test-coverage`, `linting`, `duplication`, `entropy`
  • Progress file tracking (`.taskmaster/loop-progress.txt`)
  • Tool annotations for MCP tools
  • Fallback for AI SDK versions without jsonSchema export

Full Release Notes

### Minor Changes

-   [#1571](https://github.com/eyaltoledano/claude-task-master/pull/1571) [`c2d6c18`](https://github.com/eyaltoledano/claude-task-master/commit/c2d6c18a96fce5a2d5cb50bd1ae5d58ef577501c) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - Add loop command for automated task execution with Claude Code

    **New Features:**

    -   `task-master loop` command that runs Claude Code in a Docker sandbox, executing one task per iteration based on the selected tag
    -   Built-in presets for different workflows:
        -   `default` - General task completion from the Task Master backlog
        -   `test-coverage` - Find uncovered code and write meaningful tests
        -   `linting` - Fix lint errors and type errors one by one
        -   `duplication` - Find duplicated code and refactor into shared utilities
        -   `entropy` - Find code smells and clean them up
    -   Progress file tracking to maintain context across iterations (inside `.taskmaster/loop-progress.txt`)
        -   Remember to delete this file between loops to not pollute the agent with bad context
    -   Automatic completion detection via `<loop-complete>` and `<loop-blocked>` markers

### Patch Changes

-   [#1556](https://github.com/eyaltoledano/claude-task-master/pull/1556) [`1befc6a`](https://github.com/eyaltoledano/claude-task-master/commit/1befc6a341babd825b8dd000513ffbf8a1620e62) Thanks [@TheLazyIndianTechie](https://github.com/TheLazyIndianTechie)! - fix: tolerate AI SDK versions without jsonSchema export

    Fallback to sanitized Zod schema handling when jsonSchema is unavailable, and
    align structured-output tests and registration perf thresholds to reduce CI
    failures.

    Also enforce sequential, unique subtask ids when regenerating subtasks during
    scope adjustment.

-   [#1553](https://github.com/eyaltoledano/claude-task-master/pull/1553) [`226678b`](https://github.com/eyaltoledano/claude-task-master/commit/226678b93aa01d0e62c0fac852802e9955c7ebd7) Thanks [@Crunchyman-ralph](https://github.com/Crunchyman-ralph)! - fix: Remove .default() from Zod schemas to satisfy OpenAI strict JSON schema validation

    This fixes an issue where codex-cli provider (using OpenAI API) would fail with "Missing 'dependencies'" error during task expansion. OpenAI's structured outputs require all properties to be in the 'required' array, but Zod's .default() makes fields optional. The fix removes .default() from schemas and applies defaults at the application level instead.

-   [#1543](https://github.com/eyaltoledano/claude-task-master/pull/1543) [`9a6fa1b`](https://github.com/eyaltoledano/claude-task-master/commit/9a6fa1bd2ab389097f1074fe4a4f779dee8180b6) Thanks [@triepod-ai](https://github.com/triepod-ai)! - feat: Add tool annotations for improved LLM tool understanding

    Added MCP tool annotations (readOnlyHint, destructiveHint, title) to all 12 tools to help LLMs better understand tool behavior and make safer decisions about tool execution.