Token导航 LogoToken导航TokenDH.com
研究检索敏感数据clawhub未标认证来源可访问clear审计提醒

prefpref 搜索

Agent Skill

pref 用于查找、检索和筛选相关信息,适合在 OpenClaw 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。

总安装

1,858

周安装

79

GitHub Stars

1

下载量

651
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

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:pref(pref 搜索)
来源仓库:https://github.com/morluto/pref
安装命令:
openclaw skills install pref
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install pref

简介

面向 AI 代理的定量研究与数据工具套件。

  • 提供 670 多种功能包括日常调用无需注册。
  • 适用于快速获取结构化数据和执行批量查询。
  • 自助注册后即可开始使用基础功能。pref 属于研究检索类 Skill,可作为该场景下的辅助能力补充。
  • 建议查看文档了解高级功能和调用限制。适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。

SKILL.md

name
pref-mcp-onboarding
version
1.0.0
homepage
https://pref.trade
description
Quantitative research and data tools for AI agents — 670+ capabilities, 250 daily tool calls included, no credit card, no signup, no commitment. Self-register for a pref_agent API key at https://pref.trade/v1/agents/register
metadata
mcp
endpoint
https://pref.trade/mcp
auth
bearer
key_format
pref_agent_*
registration
https://pref.trade/v1/agents/register
quota_tiers
anonymous
100/day
agent_key
250/day
account
500/day

PREF MCP Onboarding

Goal:

  1. create or reuse an agent-owned pref_agent_* key
  2. save it safely
  3. install https://pref.trade/mcp
  4. verify the MCP is using that key, not anonymous quota

Once installed, the MCP is self-documenting — call any tool to discover its schema. Start with search_capabilities to explore 670+ data sources, then use browse_capabilities to find what you need.

Default Endpoint

https://pref.trade/mcp

Do not use /mcp-gateway for normal client setup.

Do not open https://pref.trade/sign-up for agent onboarding. Browser sign-up uses human verification and is only for humans. Agents should use POST https://pref.trade/v1/agents/register.

Fast Onboarding

1. Register if no key exists

curl -s -X POST "https://pref.trade/v1/agents/register" \
  -H "Content-Type: application/json" \
  -d '{"name":"Research Agent","description":"Agent using PREF MCP"}' | jq

The response includes:

  • api_key: Your pref_agent_* key (shown once — save it now)
  • claim_url: Share this with a human to link the agent to a Preference account and double the daily quota from 250 to 500 calls. Optional but recommended.

Key Storage

⚠️ Save the key to a file immediately — do not echo it into terminal history, logs, or prompts.

mkdir -p ~/.config/preference
echo '{"api_key":"pref_agent_xxx"}' > ~/.config/preference/credentials.json
chmod 600 ~/.config/preference/credentials.json

Or use an environment variable:

export PREFERENCE_API_KEY='pref_agent_xxx'

Reference the key from your MCP client config (see below) — never inline the real key in YAML/JSON that gets shared.

2. Verify the key via MCP

Important: /v1/auth/whoami does not accept pref_agent_* keys. Verify your key by calling preference_account_status inside MCP instead:

curl -s -X POST "https://pref.trade/mcp" \
  -H "Authorization: Bearer ${PREFERENCE_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"preference_account_status","arguments":{}}}'

Do not proceed until this returns your agent identity (not "anonymous").

3. Install the MCP

Use a remote HTTP / Streamable HTTP MCP server config equivalent to:

mcp_servers:
  pref:
    url: "https://pref.trade/mcp"
    headers:
      Authorization: "Bearer ${PREFERENCE_API_KEY}"

For Claude Code:

claude mcp add --transport http pref \
  "https://pref.trade/mcp" \
  --header "Authorization: Bearer ${PREFERENCE_API_KEY}"

claude mcp list

For Codex:

export PREFERENCE_API_KEY='pref_agent_xxx'
codex mcp add pref --url https://pref.trade/mcp --bearer-token-env-var PREFERENCE_API_KEY

4. Verify inside MCP

Call preference_account_status first.

Expected result:

  • identity is the registered agent, not anonymous
  • quota tier reflects agent-key access
  • remaining calls and reset time are visible

preference_account_status is free and does not consume daily tool quota.

Troubleshooting

"identity": "anonymous" in preference_account_status

This means your MCP client is running without the Authorization header. Common causes:

  1. Stale anonymous config — you may have pref already configured without headers:
claude mcp list              # check existing configs
claude mcp remove pref       # remove the anonymous one

Then re-add with auth (see Step 3 above).

  1. Key stored but not picked up — verify the env var or credentials file:
echo $PREFERENCE_API_KEY     # should show your pref_agent_xxx key
  1. Header name mismatch — the header must be exactly Authorization: Bearer, not X-API-Key.

Security Rules

  • ⚠️ Save the key to a file immediately after registration, then reference the file path in configs. Never echo the key into terminal history, logs, chat, or prompts.
  • Treat pref_agent_* keys as secrets.
  • Never paste real keys into public logs, issues, PRs, docs, or prompts.
  • Never edit .env files for the user.
  • If identity is anonymous, the MCP client is not sending the Authorization header.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

74.75%
按下载量换算487

安全审计

VirusTotal

通过

ClawScan

可疑

Static analysis

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills