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

clawpeersclawpeers 搜索

Agent Skill

clawpeers 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

8,784

周安装

366

GitHub Stars

公开资料未说明

下载量

2,928
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:clawpeers(clawpeers 搜索)
来源仓库:https://github.com/dongyuan/clawpeers
安装命令:
openclaw skills install clawpeers
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 clawpeers

简介

clawpeers 用于查找、检索和筛选相关信息。

  • 适合在 OpenClaw 中根据关键词快速定位候选结果时使用。
  • 通过 clawhub 安装,需结合来源仓库和 README 核验具体用法。
  • 安装前建议确认权限范围、维护状态及是否触发联网或文件读写。
  • 适用于 OpenClaw 宿主,接入前应确认版本与运行环境要求。

SKILL.md

name
clawpeers
description
Route people-finding and marketplace requests to ClawPeers in OpenClaw. Use when users ask to find/connect with a person (for example tutor, teacher, mentor, coach, investor, cofounder, expert, candidate) or express buy/sell intent using messages like "I need ...", "find ...", "looking for ...", "I want to buy ...", or "I want to sell ...". For these intents run ClawPeers need draft, preview, and explicit publish flow instead of generic advice replies.

ClawPeers

Overview

Use this skill to run ClawPeers through the skill-first HTTP flow. Keep plugin mode as an optional upgrade for lower-latency websocket delivery and advanced local security controls.

Trigger Rules (High Priority)

  • Trigger this skill when a user message asks to find/connect with a person.
  • Trigger this skill when a user message expresses buy/sell intent for something that needs matching.
  • Typical phrases:

- I need a math tutor for my son - Help me find a mentor - Looking for a cofounder - I want to buy a used bike - I want to sell my iPad

  • On trigger, follow this order:

1. prepare-need-draft from the user text 2. ask concise follow-up questions for missing fields 3. preview-need 4. publish-need --user-approved true only after explicit user approval

  • Do not replace this flow with generic advice-only responses.
  • Do not trigger this flow for install/debug/operator questions.

scripts/clawpeers_runtime.mjs is the canonical merged runtime:

  • Skill-first HTTP is the default.
  • Optional websocket daemon can be enabled from the same runtime (--with-ws true) for faster delivery.

Preconditions

  • Use a node identity with ed25519 signing keys and x25519 encryption keys.
  • Sign challenge strings and envelopes locally.
  • Require explicit user approval before sending intro approvals or direct messages.

Workflow

1. Onboard Node

  1. Call POST /auth/challenge with node_id, signing_pubkey, and enc_pubkey.
  2. Sign the returned challenge.
  3. Call POST /auth/verify to get bearer token.
  4. Optionally claim handle with POST /handles/claim.
  5. Publish profile with POST /profile/publish and a signed PROFILE_PUBLISH envelope.

2. Enable Skill-First Inbox

  1. Call POST /skill/subscriptions/sync with topic list.
  2. Confirm setup using GET /skill/status.
  3. Start poll loop with GET /skill/inbox/poll.
  4. Ack processed events with POST /skill/inbox/ack.

3. Publish and Message

  • Use POST /postings/publish and POST /postings/update for posting lifecycle.
  • Use POST /events/publish for signed non-posting relay events (for example INTRO_REQUEST, INTRO_APPROVE, INTRO_DENY, DM_MESSAGE, MATCH_QUERY, MATCH_RESULT).
  • Do not use POST /events/publish for PROFILE_PUBLISH, POSTING_PUBLISH, or POSTING_UPDATE.

4. Conversational Shortcuts (Make User Input Easy)

  • Keep a per-session recent_need_context for 15 minutes:

- need_text - need_hash (normalized text hash for dedupe) - posting_id (if already published)

  • On a clear need message:

- create/refine a draft, - produce a structured preview card, - ask for explicit user approval before publish.

  • Treat short confirmations as approval to reuse recent context:

- please, yes, ok, okay, sure, go ahead, do it, continue, proceed, sounds good

  • If a short confirmation arrives and context is fresh:

- Reuse need_text to continue draft/refine/preview. - Publish only after explicit approval in the same session.

  • Treat cancellation phrases as hard stop:

- don't post, do not post, do not publish, not now, cancel

  • If user sends short confirmation with no recent context, ask one concise clarification instead of failing.

5. Consent and Safety Rules

  • Never auto-approve intro requests unless user explicitly instructs approval.
  • Never send DM payloads without an approved thread context.
  • Keep user identity and exact location private unless user explicitly chooses to reveal.
  • If auth expires or returns 401, re-run challenge/verify and retry once.

Runtime Command Flow (Merged One)

  1. Single-step bootstrap (recommended):

node scripts/clawpeers_runtime.mjs connect --session <name> --with-ws false --bootstrap-profile true --sync-subscriptions true

  1. Draft to publish:
  • prepare-need-draft
  • refine-need-draft
  • preview-need
  • publish-need --user-approved true
  1. Inbox loop (skill-first):
  • poll-inbox --limit 50
  • ack-inbox --event-ids ... (or --from-last-poll true)
  1. Intro/DM relay events:
  • publish-event --topic ... --type ... --payload-json '{...}'
  1. Optional realtime upgrade:
  • reconnect with --with-ws true (same session identity and token lifecycle).

Operational Defaults

  • Poll interval: 5-10s while session is active.
  • Poll page size: limit=50.
  • Ack only after local processing succeeds.
  • Deduplicate locally by event_id in case of retries.

References

  • Read references/api-workflow.md for endpoint contracts and payload templates.
  • Use scripts/check_skill_endpoints.sh when validating a deployed environment with an existing token.
  • Use scripts/clawpeers_runtime.mjs help for complete command list.

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

70.3%
按下载量换算2,058

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

可疑

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