x-research by rohunvora/x-research-skill
npx skills add https://github.com/rohunvora/x-research-skill --skill x-research针对 X/Twitter 的通用智能研究代理。将任何研究问题分解为针对性搜索,迭代优化,追踪线索,深入挖掘链接内容,并综合成一份附有来源的简报。
关于 X API 详情(端点、运算符、响应格式):请阅读 references/x-api.md。
所有命令均在此技能目录下运行:
cd ~/clawd/skills/x-research
source ~/.config/env/global.env
bun run x-search.ts search "<查询>" [选项]
选项:
--sort likes|impressions|retweets|recent — 排序方式(默认:likes)--since 1h|3h|12h|1d|7d — 时间过滤器(默认:最近 7 天)。也接受分钟(30m)或 ISO 时间戳。--min-likes N — 按最小点赞数过滤广告位招租
在这里展示您的产品或服务
触达数万 AI 开发者,精准高效
--min-impressions N — 按最小展示次数过滤--pages N — 要获取的页数,1-5(默认:1,100 条推文/页)--limit N — 显示的最大结果数(默认:15)--quick — 快速模式:1 页,最多 10 条结果,自动噪音过滤(-is:retweet -is:reply),1 小时缓存,成本摘要--from <用户名> — 查询中 from:用户名 的简写形式--quality — 过滤低互动推文(≥10 个赞,后处理)--no-replies — 排除回复--save — 将结果保存到 ~/clawd/drafts/x-research-{slug}-{date}.md--json — 原始 JSON 输出--markdown — 用于研究文档的 Markdown 输出除非查询已包含,否则会自动添加 -is:retweet。所有搜索都会显示预估的 API 成本。
示例:
bun run x-search.ts search "BNKR" --sort likes --limit 10
bun run x-search.ts search "from:frankdegods" --sort recent
bun run x-search.ts search "(opus 4.6 OR claude) trading" --pages 2 --save
bun run x-search.ts search "$BNKR (revenue OR fees)" --min-likes 5
bun run x-search.ts search "BNKR" --quick
bun run x-search.ts search "BNKR" --from voidcider --quick
bun run x-search.ts search "AI agents" --quality --quick
bun run x-search.ts profile <用户名> [--count N] [--replies] [--json]
获取特定用户的近期推文(默认排除回复)。
bun run x-search.ts thread <推文_id> [--pages N]
根据根推文 ID 获取完整的对话线程。
bun run x-search.ts tweet <推文_id> [--json]
bun run x-search.ts watchlist # 显示所有
bun run x-search.ts watchlist add <用户> [备注] # 添加账户
bun run x-search.ts watchlist remove <用户> # 移除账户
bun run x-search.ts watchlist check # 检查所有账户的近期动态
关注列表存储在 data/watchlist.json 中。用于心跳集成——检查关键账户是否发布了重要内容。
bun run x-search.ts cache clear # 清除所有缓存结果
15 分钟 TTL。避免重复获取相同查询。
在进行深度研究(不仅仅是快速搜索)时,请遵循此循环:
使用 X 搜索运算符将研究问题转化为 3-5 个关键词查询:
from: 特定的已知专家(broken OR bug OR issue OR migration) 等关键词(shipped OR love OR fast OR benchmark) 等关键词url:github.com 或 url: 特定域名-is:retweet(自动添加),如果需要可添加 -is:reply-airdrop -giveaway -whitelist通过 CLI 运行每个查询。每次运行后,评估:
from: 搜索的关键人物?thread 命令追踪的对话线程?web_fetch 深入挖掘的链接资源?当一条推文互动量高或是线程起始推文时:
bun run x-search.ts thread <推文_id>
当推文链接到 GitHub 仓库、博客文章或文档时,使用 web_fetch 获取。优先处理以下链接:
按主题而非查询对发现进行分组:
### [主题/发现标题]
[1-2 句话总结]
- @用户名: "[关键引述]" (NL, NI) [推文](url)
- @用户名2: "[另一个视角]" (NL, NI) [推文](url)
分享的资源:
- [资源标题](url) — [资源简介]
使用 --save 标志或手动保存到 ~/clawd/drafts/x-research-{主题-slug}-{YYYY-MM-DD}.md。
-is:reply,使用 --sort likes,缩小关键词范围OR 扩大范围,移除限制性运算符-$ -airdrop -giveaway -whitelistfrom: 或 --min-likes 50has:links 搜索在心跳检查时,可以运行 watchlist check 来查看关键账户是否发布了值得注意的内容。仅当内容真正有趣/可操作时才向 Frank 报告——不要报告常规推文。
skills/x-research/
├── SKILL.md (本文件)
├── x-search.ts (CLI 入口点)
├── lib/
│ ├── api.ts (X API 包装器:搜索、线程、资料、推文)
│ ├── cache.ts (基于文件的缓存,15分钟 TTL)
│ └── format.ts (Telegram + markdown 格式化器)
├── data/
│ ├── watchlist.json (要监控的账户)
│ └── cache/ (自动管理)
└── references/
└── x-api.md (X API 端点参考)
每周安装次数
221
仓库
GitHub 星标数
999
首次出现
2026 年 2 月 8 日
安全审计
安装于
opencode189
gemini-cli182
codex179
github-copilot173
cursor172
amp165
General-purpose agentic research over X/Twitter. Decompose any research question into targeted searches, iteratively refine, follow threads, deep-dive linked content, and synthesize into a sourced briefing.
For X API details (endpoints, operators, response format): read references/x-api.md.
All commands run from this skill directory:
cd ~/clawd/skills/x-research
source ~/.config/env/global.env
bun run x-search.ts search "<query>" [options]
Options:
--sort likes|impressions|retweets|recent — sort order (default: likes)--since 1h|3h|12h|1d|7d — time filter (default: last 7 days). Also accepts minutes (30m) or ISO timestamps.--min-likes N — filter by minimum likes--min-impressions N — filter by minimum impressions--pages N — pages to fetch, 1-5 (default: 1, 100 tweets/page)--limit N — max results to display (default: 15)--quick — quick mode: 1 page, max 10 results, auto noise filter (-is:retweet -is:reply), 1hr cache, cost summary--from <username> — shorthand for from:username in query--quality — filter low-engagement tweets (≥10 likes, post-hoc)--no-replies — exclude replies--save — save results to ~/clawd/drafts/x-research-{slug}-{date}.md--json — raw JSON output--markdown — markdown output for research docsAuto-adds -is:retweet unless query already includes it. All searches display estimated API cost.
Examples:
bun run x-search.ts search "BNKR" --sort likes --limit 10
bun run x-search.ts search "from:frankdegods" --sort recent
bun run x-search.ts search "(opus 4.6 OR claude) trading" --pages 2 --save
bun run x-search.ts search "$BNKR (revenue OR fees)" --min-likes 5
bun run x-search.ts search "BNKR" --quick
bun run x-search.ts search "BNKR" --from voidcider --quick
bun run x-search.ts search "AI agents" --quality --quick
bun run x-search.ts profile <username> [--count N] [--replies] [--json]
Fetches recent tweets from a specific user (excludes replies by default).
bun run x-search.ts thread <tweet_id> [--pages N]
Fetches full conversation thread by root tweet ID.
bun run x-search.ts tweet <tweet_id> [--json]
bun run x-search.ts watchlist # Show all
bun run x-search.ts watchlist add <user> [note] # Add account
bun run x-search.ts watchlist remove <user> # Remove account
bun run x-search.ts watchlist check # Check recent from all
Watchlist stored in data/watchlist.json. Use for heartbeat integration — check if key accounts posted anything important.
bun run x-search.ts cache clear # Clear all cached results
15-minute TTL. Avoids re-fetching identical queries.
When doing deep research (not just a quick search), follow this loop:
Turn the research question into 3-5 keyword queries using X search operators:
from: specific known experts(broken OR bug OR issue OR migration)(shipped OR love OR fast OR benchmark)url:github.com or url: specific domains-is:retweet (auto-added), add -is:reply if needed-airdrop -giveaway -whitelist if crypto topics floodingRun each query via CLI. After each, assess:
from: specifically?thread command?web_fetch?When a tweet has high engagement or is a thread starter:
bun run x-search.ts thread <tweet_id>
When tweets link to GitHub repos, blog posts, or docs, fetch with web_fetch. Prioritize links that:
Group findings by theme, not by query:
### [Theme/Finding Title]
[1-2 sentence summary]
- @username: "[key quote]" (NL, NI) [Tweet](url)
- @username2: "[another perspective]" (NL, NI) [Tweet](url)
Resources shared:
- [Resource title](url) — [what it is]
Use --save flag or save manually to ~/clawd/drafts/x-research-{topic-slug}-{YYYY-MM-DD}.md.
-is:reply, use --sort likes, narrow keywordsOR, remove restrictive operators-$ -airdrop -giveaway -whitelistfrom: or --min-likes 50has:linksOn heartbeat, can run watchlist check to see if key accounts posted anything notable. Flag to Frank only if genuinely interesting/actionable — don't report routine tweets.
skills/x-research/
├── SKILL.md (this file)
├── x-search.ts (CLI entry point)
├── lib/
│ ├── api.ts (X API wrapper: search, thread, profile, tweet)
│ ├── cache.ts (file-based cache, 15min TTL)
│ └── format.ts (Telegram + markdown formatters)
├── data/
│ ├── watchlist.json (accounts to monitor)
│ └── cache/ (auto-managed)
└── references/
└── x-api.md (X API endpoint reference)
Weekly Installs
221
Repository
GitHub Stars
999
First Seen
Feb 8, 2026
Security Audits
Gen Agent Trust HubFailSocketWarnSnykWarn
Installed on
opencode189
gemini-cli182
codex179
github-copilot173
cursor172
amp165
Python PDF处理教程:合并拆分、提取文本表格、创建PDF文件
57,000 周安装