Token导航 LogoToken导航

Agent Skill · 研究检索

seer-market(先知市场)

seer-market 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。

seer-market 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,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 70,834。

总安装 198周安装 8GitHub Stars 公开资料未说明下载 62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:seer-market(先知市场)
来源仓库:https://github.com/chris-xu0321/midaz-cli
仓库路径:skills/seer-market
安装命令:
npx skills add https://github.com/chris-xu0321/midaz-cli --skill seer-market
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/chris-xu0321/midaz-cli --skill seer-market

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,支持市场研究与数据分析。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词定位资料。通过 npx skills add 命令从指定 GitHub 仓库安装使用。安装前需确认是否有联网、文件读写或命令执行权限。

建议结合原始 README 了解数据源范围与更新机制。

来源仓库内路径:skills/seer-market

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/chris-xu0321/midaz-cli --skill seer-market;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

seer-market 怎么安装?

在终端执行 npx skills add https://github.com/chris-xu0321/midaz-cli --skill seer-market。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

seer-market 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。

seer-market 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。

seer-market 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。

seer-market 用的人多吗?

公开数据是累计安装 198 次,近一周安装 8 次,下载 62 次。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。

seer-market 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

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。
  • 第三方扫描已通过,常见的高风险行为没有命中。

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
seer-market

先知市场 · 本页

198—需要联网未明确通过

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

Seer Market Intelligence

Read seer-shared for response format, the web UI, and common rules.

Command Reference

Entity lookup

seer-q search "QUERY"           # Fuzzy search across topics, threads, assets
seer-q topic TOPIC_ID           # Topic detail: thesis, bias, all threads within it
seer-q thread THREAD_ID         # Thread detail: snapshot, all claims, evidence counts

List / browse

seer-q market                   # Global regime + all topics with thread counts
seer-q topics                   # List all topics with thread counts
seer-q threads                  # List all threads (newest activity first)
seer-q threads --topic ID       # Threads in a specific topic
seer-q threads --status active  # Filter by status (active/weakening/divided/resolved)
seer-q claims                   # Latest 100 claims (newest first)
seer-q claims --thread ID       # Claims for a specific thread
seer-q claims --status current  # Filter by status (pending/current/stale/discarded)
seer-q claims --mode observed   # Filter by mode (observed/interpreted/forecast/attributed)
seer-q sources                  # Latest 100 ingested sources
seer-q sources --decision process  # Only processed sources
seer-q sources --tier 1         # Only tier-1 sources

Snapshots

seer-q snapshot                 # Latest global regime snapshot
seer-q snapshot --history       # Regime snapshot history (default 10)
seer-q snapshot --history --limit 5  # Limit history count

Usage & audit

seer-q usage                    # Token usage summary (--since P, default 24h)
seer-q decisions                # Decision log (--stage S, --run ID, --entity-type T, --entity-id I, --limit N)

Query Strategy

Map the user's question to the right command sequence:

User intentCommands
Overall market / "how's the market"seer-q market
List all topicsseer-q topics
Hottest/most active topicseer-q topics → pick highest thread_count or most recent activity
Specific sector deep-dive (e.g., "AI infra")seer-q search "KEYWORDS" → seer-q topic ID
Specific angle/trade (e.g., "NVIDIA bear case")seer-q search "KEYWORDS" → seer-q thread ID
Analyze an asset (e.g., "analyze NVIDIA")seer-q search "ASSET" → fetch relevant topic + multiple thread details
Latest events / recent claimsseer-q claims → summarize the newest entries
Claims for a threadseer-q claims --thread ID
Recent sources / what was ingestedseer-q sources
Bull/bear case for Xseer-q search "X" → seer-q thread ID → focus on risk_case, contradicting claims
Market regime history / trendseer-q snapshot --history
Global regime detailsseer-q snapshot
Threads in a topicseer-q threads --topic ID or seer-q topic ID (which includes threads)

Key Response Fields

Market/Topics:

  • regime_summary — one-line market regime
  • standing_thesis — topic thesis
  • standing_digest — topic summary
  • bias — bullish/bearish/neutral/mixed/unclear
  • thread_count — number of threads in topic
  • view_url — link to the interactive map (always share — opens the topic on a 3D sphere the user can click around)

Threads:

  • thesis — thread thesis
  • bias, status — current stance and lifecycle
  • snapshot — detailed analysis: assessment, conviction, catalysts, outcomes, risk_case, what_breaks_it, assets_exposed, top_contradiction
  • supporting_count, contradicting_count — evidence balance
  • view_url — link to the interactive map (always share — opens the map focused on this thread in its topic context)

Claims:

  • statement — the claim text
  • claim_mode — observed/interpreted/forecast/attributed
  • thread_role — support/contradiction
  • event_date — when the event occurred
  • status — pending/current/stale/discarded
  • asset_mentions — related assets
  • source_id — link to source

Sources:

  • title, url — source identity
  • source_tier — 1 (highest) to 3
  • gate_decision — process/drop
  • published_at, ingested_at — timing

Global snapshot:

  • regime_summary — one-liner
  • snapshot.verdict — stance + rationale
  • snapshot.major_drivers — key market drivers
  • snapshot.key_uncertainties — what could change

Examples

User: "how's the market" -> seer-q market -> Summarize regime_summary, verdict, top topics by thread count -> Share the market view_url — mention they can click into topics and explore the driver network there -> Share each topic's view_url as you mention them

User: "latest 10 events" -> seer-q claims -> Take first 10 from response (already sorted newest-first) -> Summarize each claim: statement, event_date, asset_mentions -> No view_url on individual claims

User: "hottest topic right now" -> seer-q topics -> Find topic with highest thread_count or most recent thread activity -> seer-q topic ID for detail -> Summarize thesis, bias, top threads -> Share the topic view_url — the user can explore all its threads visually on the map

User: "analyze NVIDIA" -> seer-q search "NVIDIA" -> Fetch each relevant topic and thread -> Synthesize: bull case (supporting threads), bear case (contradicting/risk), key catalysts -> Include view_urls as you go so the user can click into any angle that interests them

User: "what's the bear case for AI" -> seer-q search "AI" -> Find bearish/weakening threads -> seer-q thread ID for the most relevant -> Focus on risk_case, what_breaks_it, contradicting claims -> Share thread view_url — the map shows the evidence balance visually

User: "recent sources" -> seer-q sources -> Summarize: title, tier, published_at, gate_decision -> Group by decision (processed vs dropped) if useful

User: "how has the market regime changed" -> seer-q snapshot --history -> Show regime_summary progression over time -> Note any shifts in stance

User: "what claims support thread X" -> seer-q claims --thread ID -> Filter/highlight claims with thread_role=support -> Summarize key supporting evidence -> Share thread view_url if available

平台分布

Codex

33.56%
按下载量换算21

Claude

30.54%
按下载量换算19

Cursor

18.01%
按下载量换算11

Gemini CLI

8.5%
按下载量换算5

安全审计

Gen Agent Trust Hub

通过

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。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