Token导航 LogoToken导航

Agent Skill · 研究检索

athena-package(雅典娜包)

athena-package 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

athena-package 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为需要联网;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 70,176。

总安装 212周安装 9GitHub Stars 1下载 74核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:athena-package(雅典娜包)
来源仓库:https://github.com/bluewaves-creations/bluewaves-skills
仓库路径:skills/athena-package
安装命令:
npx skills add https://github.com/bluewaves-creations/bluewaves-skills --skill athena-package
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/bluewaves-creations/bluewaves-skills --skill athena-package

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词或任务需求快速定位候选结果。它创建 Athena 笔记应用的导入包(ZIP 格式),包含 manifest.json、Markdown 笔记和可选资产文件。

使用时需遵循 aurora 标签规范组织 note 索引,支持 YAML 前置元数据和二进制资源引用,详见 references/athena-spec.md。安装前请确认是否具备 GitHub 访问权限,并评估是否会触发网络请求或令牌写入,避免安全风险。

athena-package 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/athena-package

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/bluewaves-creations/bluewaves-skills --skill a…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

athena-package 怎么安装?

在终端执行 npx skills add https://github.com/bluewaves-creations/bluewaves-skills --skill athena-pack…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

athena-package 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

athena-package 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

athena-package 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

athena-package 用的人多吗?

公开数据是累计安装 212 次,近一周安装 9 次,下载 74 次,GitHub 1 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

athena-package 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
athena-package

雅典娜包 · 本页

2121需要联网未明确有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Athena Package Creator

Create validated .athena import packages for the Athena note-taking app. These are ZIP archives containing a manifest, markdown notes with aurora tags, and optional assets.

Package Format Overview

package.athena (ZIP)
├── manifest.json              # Required: note index with aurora tags
├── notes/
│   ├── project-plan.md        # Markdown notes with optional YAML frontmatter
│   ├── meeting-notes.md
│   └── ...
└── assets/
    ├── diagram.png            # Binary assets referenced from notes
    └── ...

Read references/athena-spec.md for full schema details when handling unusual configurations.

Creation Workflow

Step 1: Gather Content

Collect all content to package:

  • From an athena-work processing session (most common)
  • From conversation results the user wants to save
  • From uploaded files the user wants to import
  • From generated content (research, summaries, analyses)

Step 2: Prepare Notes

For each note, create a markdown file with:

  1. Slug filename — lowercase, hyphenated, .md extension (e.g., project-plan.md)
  2. Mandatory header format (see below)
  3. Optional YAML frontmatter — only title and aurora are parsed; these override manifest values

Place all note files in the notes/ directory.

Mandatory Note Header Format

ALWAYS use this exact structure for every note:

# note title

> short description

---

note content here...
  • # title — Must match the manifest title for this note
  • > description — One-line summary of the note's content
  • --- — Horizontal rule separator; always present before body content
  • Body content follows in standard markdown

Example:

# Market Analysis Q1 2026

> Competitive landscape analysis covering AI assistant market trends and growth

---

## Overview

The AI assistant market grew 40% year-over-year...

Step 3: Assign Aurora Tags

Every note requires an aurora tag. Use this table for selection:

Aurora ValueUse WhenExamples
commitmentsObligations, promises, deadlines, action itemsTask lists, deadline trackers, promises made
focusDeep work, projects needing concentrationProject plans, strategy docs, specs
opsOperations, processes, recurring tasksRunbooks, checklists, SOPs
collabCollaboration, meetings, shared workMeeting notes, team decisions, feedback
lifePersonal, wellbeing, non-workPersonal goals, health notes, hobbies
exploreResearch, ideas, explorationResearch findings, brainstorms, discoveries
archiveCompleted/archived itemsFinished projects, historical records
libraryReference material, resourcesGuides, documentation, collected resources

Default selection guidance:

  • Documentation and guides: library
  • Research and analysis: explore
  • Actionable items with owners/deadlines: commitments
  • Project plans and strategy: focus
  • Meeting summaries: collab

Ask the user when the appropriate tag is ambiguous or when notes span multiple categories.

Step 4: Handle Cross-References

Note-to-note links — Use relative paths:

[Display Title](notes/other-note.md)

Asset references — Use standard markdown image/link syntax:

![Alt text](assets/filename.ext)
[Download report](assets/report.pdf)

Athena's importer remaps these paths to internal UUIDs. References that don't match any file in the package are left unchanged.

Step 5: Build manifest.json

Create the manifest with this structure:

{
  "version": 1,
  "format": "athena",
  "generatedAt": "<ISO 8601 timestamp>",
  "source": "claude",
  "notes": [
    {
      "file": "notes/slug-name.md",
      "title": "Display Title",
      "aurora": "focus",
      "assets": [
        { "name": "diagram.png", "file": "assets/diagram.png", "type": "image" }
      ]
    }
  ]
}

Required fields: version (must be 1), format (must be "athena"), notes array (at least one entry), and for each note: file, title, aurora.

Optional fields: generatedAt, source, notes[].assets (may be omitted or empty array).

Step 6: Validate, Fix, and Create

Run the validation-fix-revalidation loop:

  1. Prepare staging directory — Create a temporary directory with manifest.json, notes/, and assets/ subdirectories
  2. Validate — Run the validation script: python3 ${CLAUDE_PLUGIN_ROOT}/skills/athena-package/scripts/validate_athena_package.py <staging-dir>
  3. Fix issues — If validation reports errors, fix them:

- Missing files: create or remove from manifest - Invalid aurora tags: correct to valid values - Header format errors: add/fix # title + > description + --- - Cross-reference targets missing: fix paths or remove references

  1. Revalidate — Run validation again until clean (exit code 0)
  2. Create package — Run the creation script: python3 ${CLAUDE_PLUGIN_ROOT}/skills/athena-package/scripts/create_athena_package.py <staging-dir> [output-path]
  3. Deliver — Report the output file path and summary to the user

Validation Feedback Loop

Never skip validation. The .athena format is strict — Athena's importer rejects malformed packages. The validate-fix-revalidate cycle catches issues before the user tries to import.

Common validation errors and fixes:

ErrorFix
"manifest.json not found"Create the manifest file
"invalid aurora tag 'X'"Replace with a valid aurora value
"note file 'notes/X.md' not found"Create the file or remove from manifest
"disk file 'notes/X.md' not in manifest"Add to manifest or remove the file
"note missing header format"Add # title\n\n> description\n\n---
"cross-reference target 'notes/X.md' not found"Fix the path or create the target
"asset 'assets/X.png' not found"Add the file or remove the reference

Example

Input: 2 notes from a research session with 1 image asset.

Staging directory:

research-output/
├── manifest.json
├── notes/
│   ├── executive-summary.md
│   └── market-analysis.md
└── assets/
    └── market-chart.png

manifest.json:

{
  "version": 1,
  "format": "athena",
  "generatedAt": "2026-02-07T14:30:00Z",
  "source": "claude",
  "notes": [
    {
      "file": "notes/executive-summary.md",
      "title": "Executive Summary",
      "aurora": "focus",
      "assets": []
    },
    {
      "file": "notes/market-analysis.md",
      "title": "Market Analysis",
      "aurora": "explore",
      "assets": [
        { "name": "market-chart.png", "file": "assets/market-chart.png", "type": "image" }
      ]
    }
  ]
}

notes/executive-summary.md:

# Executive Summary

> High-level overview of Q1 2026 strategy findings

---

Based on our [Market Analysis](notes/market-analysis.md), three key themes emerge...

### Key Decisions
1. Prioritize mobile-first experience
2. Launch AI features in March

notes/market-analysis.md:

# Market Analysis

> Competitive landscape analysis for the AI assistant market

---

## Overview

![Market Share Chart](assets/market-chart.png)

The AI assistant market grew 40% YoY...

Result: research-output.athena (ZIP, 3 files, ready for import)

Error Handling

ErrorRecovery
No content to packageAsk the user what to include
Validation fails after 3 fix attemptsShow remaining errors; ask user for guidance
Script not foundReport the issue; manually create the ZIP structure
Python not availableReport the issue; manually create the ZIP
Output directory not writableTry current directory; ask user for alternative

Reference Materials

  • references/athena-spec.md — Full .athena import format specification: manifest.json schema, required/optional fields, note format with frontmatter override rules, asset handling, cross-reference syntax, aurora tag values, and import behavior (Inbox placement, deduplication, UUID remapping). Read when handling unusual package configurations or debugging import failures.

平台分布

Codex

32.9%
按下载量换算24

Claude

30.22%
按下载量换算22

Cursor

18.52%
按下载量换算14

Gemini CLI

9.75%
按下载量换算7

安全审计

Gen Agent Trust Hub

可疑

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills