X Followings Digest

Auto-fetch latest tweets from your X/Twitter followings and generate structured AI digest. Supports custom time ranges: 1 day, 3 days, 7 days, or custom. 自动抓取X/Twitter关注列表的最新推文，并使用AI分析师提示词生成结构化日报。支持自定义时间段：1天、3天、7天或自定义天数。 **Trigger Words:** - "summarize my followings", "X digest", "Twitter summary", "tweets from last 3 days", "weekly summary" **触发词：** - "总结关注列表", "X日报", "Twitter摘要", "过去3天的推文", "一周摘要" **Prerequisites:** X auth via AUTH_TOKEN & CT0 env vars

Security Scan

The skill's behavior (a shell script that needs your X/Twitter session tokens and the 'bird' CLI) matches its stated purpose, but metadata omits those sensitive requirements and the README lacks an install/secure-auth guidance — this mismatch and the use of browser session tokens is a security concern you should understand before installing.

Purpose & Capability

The skill's stated purpose (fetch followings' tweets and produce a digest) coheres with the included script and prompt template. However the registry metadata claims no required env vars or dependencies while SKILL.md and scripts explicitly require AUTH_TOKEN, CT0, and the 'bird' CLI. That omission in the declared metadata is an incoherence: the skill will not work without those sensitive items and the registry did not advertise them.

Instruction Scope

The SKILL.md and included script confine actions to fetching followings' tweets (via the 'bird' CLI) and producing JSON to be fed into the provided analyst prompt for AI summarization. There are no instructions to read unrelated files, modify other skills, or send data to arbitrary external endpoints beyond passing tweets to the agent/AI.

Install Mechanism

There is no install spec (instruction-only), and the only executable provided is a small shell script. The SKILL.md lists 'bird' CLI as a dependency but gives no install instructions; absence of an install spec is lower-risk, but the missing guidance about obtaining and verifying 'bird' (a third-party CLI) is a usability and supply-chain concern.

Credentials

The skill requires AUTH_TOKEN and CT0 (browser session tokens) according to SKILL.md and the script — these are highly sensitive (can be used to act as your account). While such tokens are functionally sufficient for scraping X without API keys, their use is sensitive and should be clearly declared; the registry metadata incorrectly lists no required env vars, causing a dangerous surprise for users. The number of credentials (two session tokens) is proportionate to the scraping approach but risky if mishandled.

Persistence & Privilege

The skill does not request persistent or global privileges: always:false, no config path writes, no self-enablement, and no system modifications in the files provided. It does recommend cron for automation (user-led), which is normal but should be set up carefully given the sensitive env vars.

What to consider before installing

Before installing or running this skill, understand: (1) SKILL.md and the included script require two browser session tokens (AUTH_TOKEN and CT0) and the third-party 'bird' CLI — these are not declared in the registry metadata, so the skill will fail without them and you may be surprised to be asked for sensitive tokens. (2) AUTH_TOKEN/CT0 are highly sensitive: anyone with them can perform actions as your account; avoid exposing them to remote/third-party services and don't store them in world-readable files. (3) Prefer safer alternatives: use official API keys with explicit, minimal scopes (if available) or create a dedicated account for scraping. (4) Inspect and verify the 'bird' CLI before installing; only run the script locally on a trusted machine. (5) If you plan to automate (cron), ensure environment variables are stored securely and the machine is trusted. If the publisher updates the registry to explicitly declare dependencies/env vars, documents how 'bird' uses the tokens, or switches to OAuth/app-key-based read-only access, that would reduce risk and could change this assessment.

X关注列表日报生成器 / X Followings Digest Generator

自动抓取你关注的人的最新推文，并生成结构化的AI日报。

Auto-fetch latest tweets from your followings and generate structured AI digest.

快速开始 / Quick Start

1. 配置X授权 / Configure X Auth

export AUTH_TOKEN="your_auth_token"
export CT0="your_ct0"

2. 获取关注列表推文 / Fetch Tweets

# 默认最近1天 / Default: last 1 day
./scripts/fetch_followings_tweets.sh

# 指定数量和时间 / Specify count & days
./scripts/fetch_followings_tweets.sh 50 1   # 50 tweets, 1 day
./scripts/fetch_followings_tweets.sh 50 3   # 50 tweets, 3 days
./scripts/fetch_followings_tweets.sh 100 7  # 100 tweets, 7 days (weekly)

3. 生成日报 / Generate Digest

将获取到的推文内容，使用 analyst_prompt_template.md 中的提示词模板进行分析。

Feed the fetched tweets to the AI using the prompt template in references/analyst_prompt_template.md.

输出格式 / Output Format

日报包含以下分类（仅显示有内容的类别）：

Digest includes (only shows categories with content):

• 🔥 重大事件 / Major Events - 具体细节和影响分析 / Specific details & impact analysis
• 🚀 产品发布 / Product Releases - 新模型、API更新、工具版本 / New models, API updates, tools
• 💡 技术洞察 / Tech Insights - 技术方案、优化技巧、代码片段 / Technical solutions, optimizations
• 🔗 资源汇总 / Resources - 论文、开源项目、教程、工具 / Papers, OSS, tutorials, tools
• 🎁 福利羊毛 / Deals & Freebies - 免费额度、优惠、赠品 / Free credits, discounts, giveaways
• 📊 舆情信号 / Signals - 争议话题、预测、警告 / Controversies, predictions, warnings

语言设置 / Language Setting

在调用AI分析时，通过提示词指定输出语言：

When calling the AI, specify output language in the prompt:

• 中文输出: 使用提示词中的 [中文] 部分
• English Output: Use the [EN] section in the prompt template
• 中英双语: 使用完整提示词，要求 bilingual output

依赖 / Dependencies

• bird CLI (X/Twitter client)
• AUTH_TOKEN & CT0 from browser cookies

注意事项 / Notes

• 推文数量越多，处理时间越长
• More tweets = longer processing time
• 建议设置定时任务每日自动运行
• Recommended: set up cron job for daily auto-run