@yottameta/yotta-skills 0.19.19 → 0.19.21
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/CHANGELOG.md +16 -0
- package/README.md +6 -6
- package/README.zh-CN.md +6 -6
- package/SKILL.md +1 -1
- package/bin/yotta-skills.js +1 -1
- package/install.sh +1 -1
- package/package.json +1 -1
- package/references/install-flow.md +1 -1
- package/references/skill-list.md +11 -4
- package/references/tutorial.md +2 -2
- package/references/walkthroughs.md +1 -1
- package/scripts/yotta-skills-mcp.py +1 -1
- package/skill-manifest.json +1 -1
- package/skills.json +37 -2
package/CHANGELOG.md
CHANGED
|
@@ -1,3 +1,19 @@
|
|
|
1
|
+
## v0.19.21 (2026-09-25)
|
|
2
|
+
|
|
3
|
+
**清单同步:元忆 0.17.1**
|
|
4
|
+
|
|
5
|
+
- `skills.json` / `references/skill-list.md` 的 yotta-memory 版本 0.17.0 → 0.17.1,承接「view 删除 AI 身份时输入错误 ID 无提示」修复。
|
|
6
|
+
- 版本对齐:package.json / SKILL.md frontmatter / skill-manifest.json / 引擎 VERSION / MCP VERSION / CHANGELOG = 0.19.21。
|
|
7
|
+
|
|
8
|
+
## v0.19.20 (2026-09-25)
|
|
9
|
+
|
|
10
|
+
**清单同步:元忆 0.17.0 + 信任层四件 + 元公 0.1.2**
|
|
11
|
+
|
|
12
|
+
- `skills.json` / `references/skill-list.md` 的 yotta-memory 版本 0.16.7 → 0.17.0;清单更新日期 → 2026-09-25。
|
|
13
|
+
- 补记信任层四件(元规 0.1.0 / 元镜 0.1.0 / 元案 0.1.0 / 元题 0.1.0)与元公 0.1.2,人工可读清单从 22 技能补齐到 27 技能,并补「合规与信任」「教育与学习」两个家族分区。
|
|
14
|
+
- 修正 `skill-list.md` 与 `skills.json` 的技能集合 / 版本一致性回归;`--list`、install、install 幂等、update 的测试计数改为读取清单长度,后续新增技能不再硬编码数量。
|
|
15
|
+
- 版本对齐:package.json / SKILL.md frontmatter / skill-manifest.json / 引擎 VERSION / MCP VERSION / CHANGELOG = 0.19.20。
|
|
16
|
+
|
|
1
17
|
## v0.19.19 (2026-09-23)
|
|
2
18
|
|
|
3
19
|
**清单同步:元忆 0.16.7(迁移口令安全 + view 根指纹)**
|
package/README.md
CHANGED
|
@@ -6,7 +6,7 @@
|
|
|
6
6
|
|
|
7
7
|
<h1 align="center">yotta-skills · 元阁 (YuanGe)</h1>
|
|
8
8
|
|
|
9
|
-
<p align="center">YottaMeta's skill-family <b>orchestration router, inventory, and one-command installer</b>: route which skills to combine for a task, then one <code>npx</code> line installs the published <code>yotta-*</code> skills (currently <b>
|
|
9
|
+
<p align="center">YottaMeta's skill-family <b>orchestration router, inventory, and one-command installer</b>: route which skills to combine for a task, then one <code>npx</code> line installs the published <code>yotta-*</code> skills (currently <b>27</b>) into any agent or directory.</p>
|
|
10
10
|
<p align="center">List the family, route a task, install everything, incrementally update, preview with <code>--dry-run</code>, pin exact versions - idempotent and zero-dependency (Node.js 18+), pure npm ecosystem.</p>
|
|
11
11
|
<p align="center">This package contains <b>no skill bodies</b> — only a manifest, a downloader, a placer and a summary. Each skill still comes from its own npm package.</p>
|
|
12
12
|
|
|
@@ -21,7 +21,7 @@
|
|
|
21
21
|
|
|
22
22
|
## What it is
|
|
23
23
|
|
|
24
|
-
Installing the whole YottaMeta family used to mean running <code>npx</code> for each skill one by one. 元阁 turns it into a single command: it reads a built-in manifest (
|
|
24
|
+
Installing the whole YottaMeta family used to mean running <code>npx</code> for each skill one by one. 元阁 turns it into a single command: it reads a built-in manifest (27 published skills), downloads each from its own npm package, places it into the target skills directory, and prints a summary (success / skipped / failed).
|
|
25
25
|
|
|
26
26
|
元阁 is also the **family orchestration planner**: given a task, check the orchestration table first — which skills to combine, in what order, and why — then install exactly those. The decision table ships in the package as <code>references/orchestration.md</code> and is summarized in <code>SKILL.md</code>.
|
|
27
27
|
|
|
@@ -134,7 +134,7 @@ Supported agent keys (17): `claude` `cursor` `codex` `gemini` `goose` `amp` `ope
|
|
|
134
134
|
|
|
135
135
|
## Installation
|
|
136
136
|
|
|
137
|
-
> **Two layers first**: `yotta-skills` (元阁) is a **family installer**, not a skill itself — it ships no skill bodies. It pulls each of the
|
|
137
|
+
> **Two layers first**: `yotta-skills` (元阁) is a **family installer**, not a skill itself — it ships no skill bodies. It pulls each of the 27 published `yotta-*` skills from its own npm package into your target directory, so **getting 元阁 is not the same as having the skills**. After you have the installer, run `install` once to actually place the whole family. A single skill (e.g. `@yottameta/yotta-memory`) is its own npm package and installs by itself; 元阁 is the **install everything at once** manager, so its install path is different.
|
|
138
138
|
|
|
139
139
|
Pick any of the four methods below; the order is the recommended priority. The package always comes from **npm** (GitHub can be slow without a proxy; npm supports mirrors). Methods 2/3 only fetch the **installer**; method 4 places only the **installer skill** into an agent's skills directory — in each case you still run `install` to bring the whole family.
|
|
140
140
|
|
|
@@ -163,7 +163,7 @@ node bin/yotta-skills.js install --dir <your-skills-dir>
|
|
|
163
163
|
|
|
164
164
|
### Method 3: GitHub Download ZIP (manual / no git)
|
|
165
165
|
|
|
166
|
-
On the GitHub repository `YottaMeta/yotta-skills`, click **Code → Download ZIP**, unzip it, then run `node bin/yotta-skills.js install --dir <your-skills-dir>`. Copying the `yotta-skills` folder into an agent's skills directory only makes the **installer skill** itself callable (it has its own `SKILL.md`) — it does not bring the
|
|
166
|
+
On the GitHub repository `YottaMeta/yotta-skills`, click **Code → Download ZIP**, unzip it, then run `node bin/yotta-skills.js install --dir <your-skills-dir>`. Copying the `yotta-skills` folder into an agent's skills directory only makes the **installer skill** itself callable (it has its own `SKILL.md`) — it does not bring the 27 skills; run `install` for that.
|
|
167
167
|
|
|
168
168
|
### Method 4: install.sh (multi-agent one-liner script)
|
|
169
169
|
|
|
@@ -173,7 +173,7 @@ bash install.sh --dir <path> # install the master installer skill to the giv
|
|
|
173
173
|
bash install.sh --list # list agents -> default directories
|
|
174
174
|
```
|
|
175
175
|
|
|
176
|
-
- `install.sh` places the **元阁 installer skill** itself so an agent can invoke the family installer. To install the
|
|
176
|
+
- `install.sh` places the **元阁 installer skill** itself so an agent can invoke the family installer. To install the 27 skills, run `node bin/yotta-skills.js install --agent <name>` (or `--dir <path>`).
|
|
177
177
|
|
|
178
178
|
### Method 5: OpenClaw / QClaw (host-native commands)
|
|
179
179
|
|
|
@@ -189,7 +189,7 @@ openclaw skills update @yottameta/yotta-skills # update
|
|
|
189
189
|
|
|
190
190
|
## Included skills
|
|
191
191
|
|
|
192
|
-
The full
|
|
192
|
+
The full 27-skill manifest with Chinese names, package names, versions and descriptions is in `references/skill-list.md` (the machine-readable source is `skills.json`). Family coverage: Security & Guardrails (12), Quality & Engineering (4), Memory & Context (3), Writing & Expression (1), Workflow (1), Entry & Guidance (1), Compliance & Trust (1), Education & Learning (4).
|
|
193
193
|
|
|
194
194
|
## How it works
|
|
195
195
|
|
package/README.zh-CN.md
CHANGED
|
@@ -7,7 +7,7 @@
|
|
|
7
7
|
<h1 align="center">yotta-skills · 元阁 (YuanGe)</h1>
|
|
8
8
|
|
|
9
9
|
<p align="center">YottaMeta 的<b>技能生态编排路由 + 全家一键安装器 + 技能盘点</b>:先按需求路由该组合哪几个技能,再一条 <code>npx</code> 命令把已发布的全部
|
|
10
|
-
<code>yotta-*</code> 技能(当前 <b>
|
|
10
|
+
<code>yotta-*</code> 技能(当前 <b>27</b> 个)装进指定智能体或目录。</p>
|
|
11
11
|
<p align="center">看清单、路由组合、装全家、增量更新、<code>--dry-run</code> 预览、<code>--pin</code> 锁版本--
|
|
12
12
|
幂等、零依赖(Node.js 18+)、纯 npm 生态。</p>
|
|
13
13
|
<p align="center">本包<b>不含任何技能本体</b>——只做「清单 + 下载 + 落位 + 汇总」;每个技能仍走各自 npm 包。</p>
|
|
@@ -23,7 +23,7 @@
|
|
|
23
23
|
|
|
24
24
|
## 这是什么
|
|
25
25
|
|
|
26
|
-
以前装元阁全家要逐个 <code>npx</code> 跑很多次。元阁把它变成一条命令:读内置清单(
|
|
26
|
+
以前装元阁全家要逐个 <code>npx</code> 跑很多次。元阁把它变成一条命令:读内置清单(27 个已发布技能)、
|
|
27
27
|
逐个从各自 npm 包下载、落位到目标技能目录、打印汇总(成功 / 跳过 / 失败)。
|
|
28
28
|
|
|
29
29
|
元阁也是全家的**编排策划层**:接到需求先查「编排策划」组合表——该组合哪几个技能、按什么顺序、各自强在哪——然后给出安装与调用**建议**,由用户确认后执行。决策表随包提供(<code>references/orchestration.md</code>),<code>SKILL.md</code> 中有摘要。
|
|
@@ -139,7 +139,7 @@ npx -y @yottameta/yotta-skills --reindex
|
|
|
139
139
|
|
|
140
140
|
## 安装
|
|
141
141
|
|
|
142
|
-
> **先分清两层**:`yotta-skills`(元阁)是「全家技能安装器」,本身**不含任何技能本体**。它所做的事是把已发布的
|
|
142
|
+
> **先分清两层**:`yotta-skills`(元阁)是「全家技能安装器」,本身**不含任何技能本体**。它所做的事是把已发布的 27 个 `yotta-*` 技能从各自 npm 包装进目标目录,所以「拿到元阁」不等于「已装齐技能」——拿到后还需运行一次 `install`,才会把全家真正放进目标目录。单个技能(如 `@yottameta/yotta-memory`)是各自独立的 npm 包、装自己即可;元阁是「一次装齐全家」的管理器,安装方式与此不同。
|
|
143
143
|
|
|
144
144
|
下面四种方式任选,顺序即推荐优先级;本包一律从 **npm** 获取(GitHub 无代理较慢,npm 支持镜像)。方式二 / 三只「拿到安装器」;方式四只把「安装器技能」本身装进智能体目录——都仍需再跑一次 `install` 才装齐全家。
|
|
145
145
|
|
|
@@ -168,7 +168,7 @@ node bin/yotta-skills.js install --dir <你的技能目录>
|
|
|
168
168
|
|
|
169
169
|
### 方式三:GitHub 下载压缩包(手动 / 无 git 环境)
|
|
170
170
|
|
|
171
|
-
在 GitHub 仓库 `YottaMeta/yotta-skills` 点 **Code → Download ZIP**,解压后跑 `node bin/yotta-skills.js install --dir <你的技能目录>` 装齐全家。把 `yotta-skills` 文件夹放进智能体技能目录只让**安装器技能**本身可被调用(它有自己的 `SKILL.md`),并不会把
|
|
171
|
+
在 GitHub 仓库 `YottaMeta/yotta-skills` 点 **Code → Download ZIP**,解压后跑 `node bin/yotta-skills.js install --dir <你的技能目录>` 装齐全家。把 `yotta-skills` 文件夹放进智能体技能目录只让**安装器技能**本身可被调用(它有自己的 `SKILL.md`),并不会把 27 个技能一并带进去——那仍需跑 `install`。
|
|
172
172
|
|
|
173
173
|
### 方式四:install.sh(多智能体一键脚本)
|
|
174
174
|
|
|
@@ -178,7 +178,7 @@ bash install.sh --dir <path> # 把「安装器技能」本身装到指定目
|
|
|
178
178
|
bash install.sh --list # 列出智能体 -> 默认目录
|
|
179
179
|
```
|
|
180
180
|
|
|
181
|
-
- `install.sh` 把「元阁安装器技能」本身装进智能体 / 目录,让代理能调用元阁;要装齐
|
|
181
|
+
- `install.sh` 把「元阁安装器技能」本身装进智能体 / 目录,让代理能调用元阁;要装齐 27 个技能,再执行 `node bin/yotta-skills.js install --agent <name>`(或 `--dir <path>`)。
|
|
182
182
|
|
|
183
183
|
### 方式五:OpenClaw / QClaw(宿主官方命令)
|
|
184
184
|
|
|
@@ -194,7 +194,7 @@ openclaw skills update @yottameta/yotta-skills # 更新
|
|
|
194
194
|
|
|
195
195
|
## 全家技能清单
|
|
196
196
|
|
|
197
|
-
|
|
197
|
+
27 个技能的 slug / 中文名 / 包名 / 版本 / 说明见 `references/skill-list.md`(机器权威源为
|
|
198
198
|
`skills.json`)。家族分布:安全与护栏(12)/ 质量与工程(4)/ 记忆与上下文(3)/ 写作与表达(1)/
|
|
199
199
|
工作流(1)/ 入口与引导(1)。
|
|
200
200
|
|
package/SKILL.md
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
---
|
|
2
2
|
name: yotta-skills
|
|
3
|
-
version: 0.19.
|
|
3
|
+
version: 0.19.21
|
|
4
4
|
description: 元阁 -- 元阁全家技能的总编排策划 + 编排路由 + 一键安装器 + 技能盘点 + 运行时 hook 适配。路由层:--route / route_request 按需求摘要给出候选组合、调用顺序、角色、置信度、依据、已装/缺失状态与安装命令,只建议不自动安装;非元阁家族已装技能按 frontmatter description 机械匹配作并列候选(标注来源与未扫描状态,只读不自动调用);策划层:按场景给出「该组合哪几个元技能、组合强在哪、怎么组合使用(安装与调用均由用户确认后执行)」;安装层:一条命令把 YottaMeta 已发布的全部 yotta-* 技能装进指定智能体或目录(默认 --pin 锁死清单精确版本);盘点层:--inventory / --reindex 扫描本机已装技能生成/更新注册表,新装技能自动被发现(install/update 后自动 re-index,会话开工只建议跑本地 --reindex;更新检查走手动 --check 或后台 --check --scheduled);运行时适配层:hook capabilities / evaluate / bind / unbind 按宿主能力矩阵执行六个统一事件并留证降级,元信 before_install 已接入安装管线;自包含零依赖,不依赖任何元技能;MCP 按需加载且需用户确认后写入配置(可选:list_installed_skills/describe_skill/reindex/route_request,不常驻,未加载降级 CLI)。支持 --list 清单 / --route 路由 / install / update / update --check(只读检查)/ update --check --scheduled(后台周检)/ update --auto(家族自动更新)/ hook 适配 / --inventory / --reindex / --dry-run 预览 / --pin(默认)/ --range。触发:需要批量安装或更新元阁全家技能、按场景组合多个元技能、路由或判断该用哪些技能、盘点或查看本机已装技能、重扫技能注册表、给某个智能体或目录一次性铺齐 yotta-* 技能、评估宿主 hook 能力、预览安装清单、锁版本安装、或用户说 元阁/装全家/一次装齐/yotta-skills/install-all/更新全家/检查更新/自动更新/hook 适配/该用哪个技能/路由技能/盘点技能/查看已装技能 等。边界(Do NOT trigger):只做「组合策划 + 静态路由建议 + 清单 + 下载 + 落位 + 汇总 + 盘点 + re-index + hook 适配」,不含技能本体、不做技能内容开发、不 -g 污染全局、不自动安装缺失技能、不静默写宿主配置或全局记忆;家族安装先自举或调用元信装前门禁,DO NOT INSTALL 阻断,非元阁家族包不自动安装。
|
|
5
5
|
license: MIT
|
|
6
6
|
metadata:
|
package/bin/yotta-skills.js
CHANGED
|
@@ -38,7 +38,7 @@ const hookAdapterLib = require('../lib/hook-adapter');
|
|
|
38
38
|
const { createInstaller, isSafeTarEntry } = require('../lib/install-pipeline');
|
|
39
39
|
|
|
40
40
|
const PKG_ROOT = path.join(__dirname, '..');
|
|
41
|
-
let VERSION = '0.19.
|
|
41
|
+
let VERSION = '0.19.21';
|
|
42
42
|
try { VERSION = require(path.join(PKG_ROOT, 'package.json')).version; } catch (_) { /* keep fallback */ }
|
|
43
43
|
|
|
44
44
|
function loadManifest() {
|
package/install.sh
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
#!/usr/bin/env bash
|
|
2
2
|
# yotta-skills(元阁)多智能体安装脚本(YottaSkills)
|
|
3
3
|
# 用途:把「元阁安装器技能」本身装进智能体/目录(它自带 SKILL.md,让代理能调用元阁)。
|
|
4
|
-
# 注意:这**不是**把
|
|
4
|
+
# 注意:这**不是**把 27 个 yotta-* 技能一并装齐——装全家需再跑
|
|
5
5
|
# `node bin/yotta-skills.js install --agent <name>`(或 --dir <path>)。
|
|
6
6
|
# 用法:
|
|
7
7
|
# bash install.sh --agent <name> # 把安装器技能装到智能体默认用户级目录
|
package/package.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"name": "@yottameta/yotta-skills",
|
|
3
|
-
"version": "0.19.
|
|
3
|
+
"version": "0.19.21",
|
|
4
4
|
"description": "YuanGe (元阁) - orchestration routing, inventory, and one-command installer for the YottaMeta skill family (zero dependencies, Node.js only).",
|
|
5
5
|
"license": "MIT",
|
|
6
6
|
"keywords": [
|
package/references/skill-list.md
CHANGED
|
@@ -1,14 +1,14 @@
|
|
|
1
|
-
# 全家技能清单(
|
|
1
|
+
# 全家技能清单(27 个)
|
|
2
2
|
|
|
3
3
|
> 机器权威源:`skills.json`(随包分发);本文件为人工可读副本,新技能发布后须同步更新
|
|
4
|
-
> `skills.json`(登记表为内部台账,不随包分发)。清单更新日期:2026-09-
|
|
4
|
+
> `skills.json`(登记表为内部台账,不随包分发)。清单更新日期:2026-09-25。
|
|
5
5
|
|
|
6
6
|
| slug | 中文名 | npm 包 | 清单版本 | 说明 |
|
|
7
7
|
|---|---|---|---|---|
|
|
8
8
|
| `yotta-anti-shallow` | 元谨 | `@yottameta/yotta-anti-shallow` | 1.4.0 | 防 AI 敷衍规则引擎(深入分析/根因追溯时激活) |
|
|
9
9
|
| `yotta-code-quality` | 元质 | `@yottameta/yotta-code-quality` | 0.4.1 | 结对式代码质量评审(十二类腐化风险 + 0-100 健康分) |
|
|
10
10
|
| `yotta-workflow` | 元序 | `@yottameta/yotta-workflow` | 0.4.1 | 跨会话/跨项目通用工作流协议(状态就近存) |
|
|
11
|
-
| `yotta-memory` | 元忆 | `@yottameta/yotta-memory` | 0.
|
|
11
|
+
| `yotta-memory` | 元忆 | `@yottameta/yotta-memory` | 0.17.1 | 有权限边界的文件式智能体记忆(FACT/PREF/BOUND/COMMIT) |
|
|
12
12
|
| `yotta-learn` | 元习 | `@yottameta/yotta-learn` | 0.2.1 | 学习沉淀 CLI:错误/纠正/洞见沉淀为 .learnings/ 条目 |
|
|
13
13
|
| `yotta-security-audit` | 元安 | `@yottameta/yotta-security-audit` | 0.2.4 | 安全扫描引擎:13 类检测器 + 系统安全基线 |
|
|
14
14
|
| `yotta-vetter` | 元审 | `@yottameta/yotta-vetter` | 0.2.5 | 安全审查协议:四阶段 review + SAFE TO INSTALL 判定 |
|
|
@@ -27,6 +27,11 @@
|
|
|
27
27
|
| `yotta-triage` | 元鉴 | `@yottameta/yotta-triage` | 0.1.3 | 恶意样本静态初筛引擎(哈希/熵/字符串/PE-ELF) |
|
|
28
28
|
| `yotta-prompt` | 元引 | `@yottameta/yotta-prompt` | 0.2.2 | 意图澄清 + 生态入口(常驻注入,map 串联元阁全家) |
|
|
29
29
|
| `yotta-verify` | 元信 | `@yottameta/yotta-verify` | 0.3.2 | 装前安全扫描器 + audited 徽章(prompt injection + 危险模式) |
|
|
30
|
+
| `yotta-compliance` | 元规 | `@yottameta/yotta-compliance` | 0.1.0 | 本地确定性合规条款审查(PIPL + 数据出境规则包 / 证据锚点 / 框架覆盖分级) |
|
|
31
|
+
| `yotta-mirror` | 元镜 | `@yottameta/yotta-mirror` | 0.1.0 | 本地确定性学情分析(知识点掌握率 / 分层与临界生 / 薄弱点诊断,学生数据不出本机) |
|
|
32
|
+
| `yotta-lesson` | 元案 | `@yottameta/yotta-lesson` | 0.1.0 | 本地确定性教案骨架与覆盖校验(课标映射 / 环节时间精确分配 / 缺口检查) |
|
|
33
|
+
| `yotta-item` | 元题 | `@yottameta/yotta-item` | 0.1.0 | 本地确定性组卷与试卷校验(双向细目表 / 题库选题 / 缺口检查) |
|
|
34
|
+
| `yotta-school-doc` | 元公 | `@yottameta/yotta-school-doc` | 0.1.2 | 本地确定性学校公文可信起草与校验(文种规则包 / 可复算骨架 / 12 类缺口) |
|
|
30
35
|
|
|
31
36
|
## 家族分布
|
|
32
37
|
|
|
@@ -36,5 +41,7 @@
|
|
|
36
41
|
- 写作与表达(1):元真
|
|
37
42
|
- 工作流(1):元序
|
|
38
43
|
- 入口与引导(1):元引
|
|
44
|
+
- 合规与信任(1):元规
|
|
45
|
+
- 教育与学习(4):元镜 / 元案 / 元题 / 元公
|
|
39
46
|
|
|
40
|
-
合计
|
|
47
|
+
合计 27 个已发布技能;本安装器自身(yotta-skills / 元阁)归「分发与安装」家族。
|
package/references/tutorial.md
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
# 元阁中文教程(新手全流程)
|
|
2
2
|
|
|
3
3
|
> 配套技能:元阁 yotta-skills(全家技能一键安装 CLI,依赖 Node.js 18+ / npm / 系统 tar)。
|
|
4
|
-
> 目标:把元阁已发布的
|
|
4
|
+
> 目标:把元阁已发布的 27 个 `yotta-*` 技能一次装进指定智能体或目录,并学会预览、更新、
|
|
5
5
|
> 锁版本与常见排查。
|
|
6
6
|
|
|
7
7
|
## 1. 教程目标与前置
|
|
@@ -18,7 +18,7 @@
|
|
|
18
18
|
npx -y @yottameta/yotta-skills --list
|
|
19
19
|
```
|
|
20
20
|
|
|
21
|
-
输出
|
|
21
|
+
输出 27 个技能:slug / 中文名 / 包名与版本范围 / 清单版本 / 一句话说明。
|
|
22
22
|
确认某个技能在不在清单里,可加技能名过滤:
|
|
23
23
|
|
|
24
24
|
```bash
|
package/skill-manifest.json
CHANGED
package/skills.json
CHANGED
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
{
|
|
2
2
|
"manifestVersion": 1,
|
|
3
|
-
"updated": "2026-09-
|
|
3
|
+
"updated": "2026-09-25",
|
|
4
4
|
"note": "清单同步要求:每次技能发布新版本后,按 YottaSkills 技能登记表更新本文件(登记表为内部台账,不随包分发)。",
|
|
5
5
|
"skills": [
|
|
6
6
|
{
|
|
@@ -28,7 +28,7 @@
|
|
|
28
28
|
"slug": "yotta-memory",
|
|
29
29
|
"name": "元忆",
|
|
30
30
|
"pkg": "@yottameta/yotta-memory",
|
|
31
|
-
"version": "0.
|
|
31
|
+
"version": "0.17.1",
|
|
32
32
|
"desc": "有权限边界的文件式智能体记忆(FACT/PREF/BOUND/COMMIT)"
|
|
33
33
|
},
|
|
34
34
|
{
|
|
@@ -156,6 +156,41 @@
|
|
|
156
156
|
"pkg": "@yottameta/yotta-verify",
|
|
157
157
|
"version": "0.3.2",
|
|
158
158
|
"desc": "装前安全扫描器 + audited 徽章(prompt injection + 危险模式)"
|
|
159
|
+
},
|
|
160
|
+
{
|
|
161
|
+
"slug": "yotta-compliance",
|
|
162
|
+
"name": "元规",
|
|
163
|
+
"pkg": "@yottameta/yotta-compliance",
|
|
164
|
+
"version": "0.1.0",
|
|
165
|
+
"desc": "本地确定性合规条款审查(PIPL + 数据出境规则包 / 证据锚点 / 框架覆盖分级)"
|
|
166
|
+
},
|
|
167
|
+
{
|
|
168
|
+
"slug": "yotta-mirror",
|
|
169
|
+
"name": "元镜",
|
|
170
|
+
"pkg": "@yottameta/yotta-mirror",
|
|
171
|
+
"version": "0.1.0",
|
|
172
|
+
"desc": "本地确定性学情分析(知识点掌握率 / 分层与临界生 / 薄弱点诊断,学生数据不出本机)"
|
|
173
|
+
},
|
|
174
|
+
{
|
|
175
|
+
"slug": "yotta-lesson",
|
|
176
|
+
"name": "元案",
|
|
177
|
+
"pkg": "@yottameta/yotta-lesson",
|
|
178
|
+
"version": "0.1.0",
|
|
179
|
+
"desc": "本地确定性教案骨架与覆盖校验(课标映射 / 环节时间精确分配 / 缺口检查)"
|
|
180
|
+
},
|
|
181
|
+
{
|
|
182
|
+
"slug": "yotta-item",
|
|
183
|
+
"name": "元题",
|
|
184
|
+
"pkg": "@yottameta/yotta-item",
|
|
185
|
+
"version": "0.1.0",
|
|
186
|
+
"desc": "本地确定性组卷与试卷校验(双向细目表 / 题库选题 / 缺口检查)"
|
|
187
|
+
},
|
|
188
|
+
{
|
|
189
|
+
"slug": "yotta-school-doc",
|
|
190
|
+
"name": "元公",
|
|
191
|
+
"pkg": "@yottameta/yotta-school-doc",
|
|
192
|
+
"version": "0.1.2",
|
|
193
|
+
"desc": "本地确定性学校公文可信起草与校验(文种规则包 / 可复算骨架 / 12 类缺口)"
|
|
159
194
|
}
|
|
160
195
|
]
|
|
161
196
|
}
|