flavor-code 1.2.14 → 1.2.15
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +8 -3
- package/README.zh-CN.md +8 -3
- package/dist/agent/planner.d.ts +2 -0
- package/dist/{app-UJAGMKD5.js → app-3KW2FTVJ.js} +1 -1
- package/dist/{chunk-4YXDGGIC.js → chunk-5TG3E2EJ.js} +561 -74
- package/dist/{chunk-KL6TR2KG.js → chunk-BKARQJQZ.js} +1 -1
- package/dist/cli.js +78 -16
- package/dist/desktop/main.js +875 -328
- package/dist/desktop-renderer/assets/{index-CcD0D6x0.js → index-BCJjMvbX.js} +1 -1
- package/dist/desktop-renderer/index.html +1 -1
- package/dist/evolve/store.d.ts +26 -0
- package/dist/git/insights.d.ts +52 -0
- package/dist/git/service.d.ts +43 -0
- package/dist/git/tools.d.ts +9 -0
- package/dist/sdk/index.js +2 -2
- package/dist/session/store.d.ts +1 -0
- package/dist/ui/commands.d.ts +8 -2
- package/dist/ui/session.d.ts +2 -0
- package/package.json +1 -1
package/README.md
CHANGED
|
@@ -32,12 +32,13 @@ Flavor Code connects to OpenAI, Anthropic, or compatible services and works with
|
|
|
32
32
|
| | Capability | What you get |
|
|
33
33
|
| --- | --- | --- |
|
|
34
34
|
| 🖥️ | **One runtime, three entry points** | CLI, Electron, and VS Code share model configuration, sessions, and tooling |
|
|
35
|
-
| 🧭 | **Controlled progress on complex tasks** | Task plans, sub-agents, steering, follow-ups, `/loop`,
|
|
35
|
+
| 🧭 | **Controlled progress on complex tasks** | Task plans, sub-agents, steering, follow-ups, `/loop`, `/goal`, and conflict-safe parallel execution (tasks owning overlapping files run serially) |
|
|
36
36
|
| ⏪ | **Traceable, resumable results** | Full timeline, checkpoints, rewind, traces, diffs, and failure audits |
|
|
37
37
|
| 🧠 | **Local long-term context** | Memory, Skills, plugins, and project guides stored on your machine |
|
|
38
38
|
| 🔎 | **Code graph navigation** | A local AST code-graph index (`.flavor/astgraph/`) powers `ast_search`/`ast_callers`/`ast_impact` queries for precise symbol lookup and reachability tracing |
|
|
39
|
+
| 🌿 | **Git-native workflows** | `/commit` drafts a Conventional-Commits message for staged changes and commits after confirmation; `/review` audits uncommitted changes; the read-only `GitHistory` tool explains when and why code changed |
|
|
39
40
|
| 🎨 | **E2E requirement-to-delivery** | From a rough requirement or a design export to a delivered product: PRD, interactive prototype, visual implementation, API integration, autonomous acceptance, and scored delivery (Electron only) |
|
|
40
|
-
| 🔁 | **Bounded self-improvement** | Repeated tool failures are captured, deduped, and proposed as suggestions; fixes ship as plugins
|
|
41
|
+
| 🔁 | **Bounded self-improvement** | Repeated tool failures are captured, deduped, and proposed as suggestions; fixes ship as sandbox-verified plugins or as learned guardrail rules injected into future prompts, with run trends and rule management (`/evolve`) |
|
|
41
42
|
| 🛡️ | **Clear permission boundaries** | Independent control over read, write, Shell, network, and destructive actions; Docker supported |
|
|
42
43
|
|
|
43
44
|
## Quick Start
|
|
@@ -155,12 +156,16 @@ Common commands:
|
|
|
155
156
|
| `/mcp` | View and manage MCP servers |
|
|
156
157
|
| `/loop <goal>` | Run an autonomous loop with verification |
|
|
157
158
|
| `/goal <objective>` | Run the plan, execute, adversarial-review workflow |
|
|
158
|
-
| `/
|
|
159
|
+
| `/commit [hint]` | Draft a Conventional-Commits message for staged changes and commit after confirmation |
|
|
160
|
+
| `/review [focus]` | Review uncommitted changes for bugs and risks before committing |
|
|
161
|
+
| `/evolve <signals\|suggest\|improve ...>` | Self-improvement loop: review repeated tool failures, scaffold fix plugins, manage run trends and learned guardrail rules, verify and hot-reload |
|
|
159
162
|
| `/pals`, `/chat`, `/co-work` | Discover and collaborate with other local CLI instances |
|
|
160
163
|
| `/audit` | View tool failure audits |
|
|
161
164
|
|
|
162
165
|
You can submit steering or queue follow-ups while a run is in progress; once the current model response finishes, the task picks up new instructions at safe boundaries.
|
|
163
166
|
|
|
167
|
+
`/commit` and `/review` use the cheap sub-agent model and degrade gracefully when it is unavailable. Session checkpoints are tagged with the current git state (`branch@sha`), so `/tree` shows what the workspace looked like at each node.
|
|
168
|
+
|
|
164
169
|
#### CLI pals and cross-project work
|
|
165
170
|
|
|
166
171
|
Interactive CLI instances on the same Windows or macOS user account can collaborate over local-only IPC (Windows named pipes or Unix sockets; no TCP fallback). Give each window a memorable alias:
|
package/README.zh-CN.md
CHANGED
|
@@ -32,12 +32,13 @@ Flavor Code 接入 OpenAI、Anthropic 或兼容服务,在受控工作区内使
|
|
|
32
32
|
| | 能力 | 你得到什么 |
|
|
33
33
|
| --- | --- | --- |
|
|
34
34
|
| 🖥️ | **一个运行时,三个入口** | CLI、Electron 与 VS Code 共享模型配置、会话和工具能力 |
|
|
35
|
-
| 🧭 | **复杂任务可控推进** | 任务计划、子 Agent、steering、follow-up、`/loop
|
|
35
|
+
| 🧭 | **复杂任务可控推进** | 任务计划、子 Agent、steering、follow-up、`/loop`、`/goal`,并行任务自动避免写冲突(拥有重叠文件的任务串行执行) |
|
|
36
36
|
| ⏪ | **结果可追溯、可恢复** | 完整时间线、checkpoint、rewind、trace、Diff 和失败审计 |
|
|
37
37
|
| 🧠 | **本地长期上下文** | 记忆、Skill、插件和项目指南均保存在本机 |
|
|
38
38
|
| 🔎 | **代码图导航** | 本地 AST 代码图索引(`.flavor/astgraph/`),通过 `ast_search`/`ast_callers`/`ast_impact` 等查询精确定位符号、追踪可达性 |
|
|
39
|
+
| 🌿 | **Git 原生工作流** | `/commit` 为暂存改动生成 Conventional Commits 提交信息并确认提交;`/review` 审查未提交改动;只读 `GitHistory` 工具回答“这段代码为什么是这样” |
|
|
39
40
|
| 🎨 | **E2E 需求到交付** | 从粗需求或设计稿到可交付产品:PRD、交互原型、视觉还原、接口联调、自主验收与评分交付(仅 Electron) |
|
|
40
|
-
| 🔁 | **有界自进化** |
|
|
41
|
+
| 🔁 | **有界自进化** | 重复的工具失败被捕获、去重并形成建议;修复以沙箱验证过的插件形式落地,或沉淀为注入后续提示词的 guardrail 规则,并支持运行趋势与规则管理(`/evolve`) |
|
|
41
42
|
| 🛡️ | **明确的权限边界** | 分别控制读、写、Shell、网络和破坏性操作,也可使用 Docker |
|
|
42
43
|
|
|
43
44
|
## 快速开始
|
|
@@ -155,12 +156,16 @@ OAuth PKCE 的运行时行为与配置约定见 [PKCE 规范](./docs/specs/pkce-
|
|
|
155
156
|
| `/mcp` | 查看和管理 MCP 服务 |
|
|
156
157
|
| `/loop <goal>` | 运行带验证的自治循环 |
|
|
157
158
|
| `/goal <objective>` | 运行规划、执行、对抗审查流程 |
|
|
158
|
-
| `/
|
|
159
|
+
| `/commit [hint]` | 为暂存改动生成 Conventional Commits 提交信息,确认后提交 |
|
|
160
|
+
| `/review [focus]` | 提交前审查未提交改动的缺陷与风险 |
|
|
161
|
+
| `/evolve <signals\|suggest\|improve ...>` | 自进化循环:查看重复工具失败、脚手架修复插件、管理运行趋势与 guardrail 规则、验证并热重载 |
|
|
159
162
|
| `/pals`、`/chat`、`/co-work` | 发现并协作其他本机 CLI 实例 |
|
|
160
163
|
| `/audit` | 查看工具失败审计 |
|
|
161
164
|
|
|
162
165
|
运行中可以提交 steering 或排队 follow-up;当前模型响应结束后,任务会在安全边界处接收新指令。
|
|
163
166
|
|
|
167
|
+
`/commit` 与 `/review` 使用廉价子 Agent 模型,模型不可用时优雅降级。会话 checkpoint 会标记当前 git 状态(`branch@sha`),`/tree` 可以看到每个节点对应的工作区现场。
|
|
168
|
+
|
|
164
169
|
#### CLI Pals 与跨项目协作
|
|
165
170
|
|
|
166
171
|
同一 Windows 或 macOS 用户下的交互式 CLI 可以通过纯本地 IPC 协作(Windows named pipe 或 Unix socket,不回退到 TCP)。先给每个窗口一个容易识别的别名:
|
package/dist/agent/planner.d.ts
CHANGED
|
@@ -6,6 +6,7 @@ export declare const TaskNodeSchema: z.ZodObject<{
|
|
|
6
6
|
dependencies: z.ZodArray<z.ZodString>;
|
|
7
7
|
expectedOutputs: z.ZodArray<z.ZodString>;
|
|
8
8
|
verification: z.ZodArray<z.ZodString>;
|
|
9
|
+
files: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
9
10
|
}, z.core.$strict>;
|
|
10
11
|
export declare const TaskGraphSchema: z.ZodObject<{
|
|
11
12
|
nodes: z.ZodArray<z.ZodObject<{
|
|
@@ -14,6 +15,7 @@ export declare const TaskGraphSchema: z.ZodObject<{
|
|
|
14
15
|
dependencies: z.ZodArray<z.ZodString>;
|
|
15
16
|
expectedOutputs: z.ZodArray<z.ZodString>;
|
|
16
17
|
verification: z.ZodArray<z.ZodString>;
|
|
18
|
+
files: z.ZodOptional<z.ZodArray<z.ZodString>>;
|
|
17
19
|
}, z.core.$strict>>;
|
|
18
20
|
}, z.core.$strict>;
|
|
19
21
|
export type TaskNode = z.infer<typeof TaskNodeSchema>;
|