v0.5.0

v0.5.0 Release Notes

Summary

Refactored the project into a pnpm monorepo architecture, updated all README documentation, and bumped version to 0.5.0.

🔄 Changed

• Monorepo Architecture (de4718c): Converted to pnpm monorepo with packages/cli (published as learn-anything-cli) and packages/gui (private, future GUI). Build, test, and lint commands now support per-package execution via pnpm -F.
• Simplified Build (c073a13): Replaced the custom build.js wrapper with direct tsc compilation, reducing indirection and making the build process more standard.

✨ Added

• README Enhancements (26b15ea): Added badges, monorepo structure diagram, and footer to all READMEs for better visuals and discoverability.

---

v0.5.0 发布说明

简介

将项目重构为 pnpm monorepo 架构，更新所有 README 文档，并升级版本号至 0.5.0。

🔄 变更

• Monorepo 架构 (de4718c): 转换为 pnpm monorepo，包含 packages/cli（发布为 learn-anything-cli）和 packages/gui（私有，未来的 GUI）。构建、测试、lint 命令现支持通过 pnpm -F 按包执行。
• 简化构建流水线 (c073a13): 移除自定义 build.js 包装器，直接使用 tsc 编译，减少间接层，使构建过程更标准化。

✨ 新增

• README 增强 (26b15ea): 所有 README 添加 badges、monorepo 结构图和 footer，提升视觉效果和可发现性。

v0.4.2

v0.4.2

Fixed

• YAML frontmatter compatibility for Codex: Changed Dual-mode: to Dual-mode (...) in SKILL_DESCRIPTION to avoid colon being interpreted as a YAML key-value separator causing invalid frontmatter in Codex.

---

修复

• Codex YAML frontmatter 兼容性: 将 SKILL_DESCRIPTION 中的 Dual-mode: 改为 Dual-mode (...)，避免冒号被 YAML 解析器视为键值分隔符导致 Codex 中 frontmatter 非法。

v0.4.1

Added

• Directory-based explanation storage: Explanation sessions now store files organized by directory structure, matching the topic hierarchy in state.json for better session management and navigation.

新增

• 基于目录的解释存储：解释会话现在会按目录结构存储文件，与 state.json 中的主题层级结构相匹配，从而更好地管理和导航会话。

v0.4.0

Added

• Learn Protocol v1: state.json is now the single source of truth for all learning data, using a hierarchical knowledge map format (domains → concepts → details). The old dual-file model (state.yaml + hand-written knowledge-map.md) is replaced — knowledge-map.md is now a generated artifact produced by render.mjs from state.json, never edited directly. AI instructions explicitly forbid reading or writing knowledge-map.md as a data source.
• Automatic v0→v1 migration: Existing learning data is auto-migrated on learn-anything init or update, with backup files created for safety.
• Schema validation: render.mjs validates state.json against the v1 schema before generating knowledge-map.md, with clear error messages on field mismatches.
• Status script: New standalone status.mjs script reads state.json and outputs a formatted heatmap or topic summary, reducing AI token spend. Supports --locale en|zh-CN for i18n output.
• Shared utils (utils.mjs): Extracted shared types, validation, and helpers used by both render.mjs and status.mjs.

Changed

• Prompt compression: Reduced skill template INSTRUCTIONS by ~69% (457 fewer lines) across 4 workflow templates, eliminating redundancy while preserving all functional behavior.
• Unified learning icons: Replaced mixed icon styles with a consistent colored circle set (🟢 🔵 🟠 ⚪) across render.mjs, migration code, skill templates, and test fixtures.
• All 5 workflow templates updated to read/write state.json only, dropping knowledge-map.md as a data source.

新增

• learn protocol v1：state.json 现在是所有学习数据的唯一数据源，采用层级式知识图谱格式（领域 → 概念 → 细节）。旧的双文件模型（state.yaml + 手写knowledge-map.md）已被替换——knowledge-map.md 现在是由 render.mjs 根据 state.json 生成的，不会直接编辑。AI 指令明确禁止读取或写入 knowledge-map.md 作为数据源。

• v0→v1 自动迁移：现有learn数据会在 learn-anything init 或 update 时自动迁移，并创建备份文件以确保安全。

• schema校验：render.mjs 会在生成 knowledge-map.md 之前，根据 v1 模式校验 state.json，并在字段不匹配时显示清晰的错误信息。

• status脚本：新增独立的 status.mjs 脚本，读取 state.json 文件并输出格式化的热图或主题摘要，从而降低token消耗。支持 --locale en|zh-CN 参数， 暂时只支持中英文输出。

• utils (utils.mjs)：提取了 render.mjs 和 status.mjs 共用的类型、校验和辅助函数。

变更

• prompt压缩：在 4 个工作流模板中，技能模板的 INSTRUCTIONS 代码减少了约 69%（减少了 457 行），消除了冗余，同时保留了所有功能。

• 统一学习图标：在 render.mjs、迁移代码、技能模板和测试用例中，将混合的图标样式替换为统一的彩色圆形图标集 (🟢 🔵 🟠 ⚪)。

• 所有 5 个工作流模板 已更新为仅读取/写入 state.json，不再使用 knowledge-map.md 作为数据源。

v0.3.1

Added

• Optional Context7 MCP integration for documentation verification during init. When enabled, generated skill files (topic, explain, practice) include guidance instructing the AI to verify explanations against official documentation using Context7 MCP tools (resolve-library-id + query-docs).
• --context7 / --no-context7 CLI flags for non-interactive Context7 control.
• After init, displays a setup hint with a link to Context7 docs for manual MCP configuration.
• i18n support for Context7 prompts in both en and zh-CN.

新增

• 可选的 Context7 MCP 集成，用于在 init 过程中进行文档验证。启用后，生成的技能文件（主题、解释、练习）将包含指导，指示 AI 使用 Context7 MCP 工具（resolve-library-id + query-docs）根据官方文档验证解释。

• --context7 / --no-context7 CLI 标志，用于非交互式 Context7 控制。

• 初始化后，显示设置提示，其中包含指向 Context7 文档的链接，以便手动配置 MCP。

• 支持 Context7 提示的国际化，支持 en 和 zh-CN 两种语言。

v0.4.0.beta.1

Added

• Learn Protocol v1: state.json is now the single source of truth for all learning data, using a hierarchical knowledge map format (domains → concepts → details). The old dual-file model (state.yaml + hand-written knowledge-map.md) is replaced — knowledge-map.md is now a generated artifact produced by render.mjs from state.json, never edited directly. AI instructions explicitly forbid reading or writing knowledge-map.md as a data source.
• Automatic v0→v1 migration: Existing learning data is auto-migrated on learn-anything init or update, with backup files created for safety.
• Schema validation: render.mjs validates state.json against the v1 schema before generating knowledge-map.md, with clear error messages on field mismatches.
• Status script: New standalone status.mjs script reads state.json and outputs a formatted heatmap or topic summary, reducing AI token spend. Supports --locale en|zh-CN for i18n output.
• Shared utils (utils.mjs): Extracted shared types, validation, and helpers used by both render.mjs and status.mjs.

Changed

• Prompt compression: Reduced skill template INSTRUCTIONS by ~69% (457 fewer lines) across 4 workflow templates, eliminating redundancy while preserving all functional behavior.
• Unified learning icons: Replaced mixed icon styles with a consistent colored circle set (🟢 🔵 🟠 ⚪) across render.mjs, migration code, skill templates, and test fixtures.
• All 5 workflow templates updated to read/write state.json only, dropping knowledge-map.md as a data source.

新增

• learn protocol v1：state.json 现在是所有学习数据的唯一数据源，采用层级式知识图谱格式（领域 → 概念 → 细节）。旧的双文件模型（state.yaml + 手写knowledge-map.md）已被替换——knowledge-map.md 现在是由 render.mjs 根据 state.json 生成的，不会直接编辑。AI 指令明确禁止读取或写入 knowledge-map.md 作为数据源。

• v0→v1 自动迁移：现有learn数据会在 learn-anything init 或 update 时自动迁移，并创建备份文件以确保安全。

• schema校验：render.mjs 会在生成 knowledge-map.md 之前，根据 v1 模式校验 state.json，并在字段不匹配时显示清晰的错误信息。

• status脚本：新增独立的 status.mjs 脚本，读取 state.json 文件并输出格式化的热图或主题摘要，从而降低token消耗。支持 --locale en|zh-CN 参数， 暂时只支持中英文输出。

• utils (utils.mjs)：提取了 render.mjs 和 status.mjs 共用的类型、校验和辅助函数。

变更

• prompt压缩：在 4 个工作流模板中，技能模板的 INSTRUCTIONS 代码减少了约 69%（减少了 457 行），消除了冗余，同时保留了所有功能。

• 统一学习图标：在 render.mjs、迁移代码、技能模板和测试用例中，将混合的图标样式替换为统一的彩色圆形图标集 (🟢 🔵 🟠 ⚪)。

• 所有 5 个工作流模板 已更新为仅读取/写入 state.json，不再使用 knowledge-map.md 作为数据源。

v0.3.0

Added

• CI workflow (GitHub Actions): lint, test, and build on every push and PR.
• Pre-commit hooks: Husky with lint-staged for ESLint, Prettier, and commitlint.

Fixed

• Session files now written BEFORE echoing to conversation in learn:practice and learn:explain workflows, eliminating drift between saved content and chat output.
• Test path assertions made cross-platform compatible (Windows vs Unix path separators).

v0.2.1

Fixed

• Relax engines.node from >=20.19.0 to >=20.0.0 for broader compatibility.

v0.2.0

Added

• Dual-mode practice: Project Mode creates real code files in your IDE; Chat Mode for conceptual discussion.
• Persist learning session records for continuity across sessions.

Fixed

• Reword session-save timing from "after" to "in the same turn" for clarity.

v0.1.0

Added

• learn-anything CLI: generate skill and command files for 30+ AI coding tools.
• init command: interactive tool detection and selection, skill generation.
• update command: update existing skill files.
• Five learning workflows: topic, explain, practice, review, status.
• Locale support: English (en) and Chinese (zh-CN).
• MIT License.