Token导航 LogoToken导航

Agent Skill · 研究检索

agenttrust(agenttrust 搜索)

agenttrust 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。

agenttrust 是研究检索类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为需要联网;许可证为 MIT-0,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 8,859。

总安装 15,429周安装 656GitHub Stars 1下载 5,405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
MIT-0
版本
1.0.6

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:agenttrust(agenttrust 搜索)
来源仓库:https://github.com/agenttrust/agenttrust
安装命令:
openclaw skills install agenttrust
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 agenttrust

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。

agenttrust 为代理提供电子邮件、文件存储与即时消息通信能力。适合构建具备自主沟通与协作功能的智能代理体系。以 your-agent@agenttrust.ai 身份发送邮件与聊天消息。通过 clawhub 安装,需绑定有效邮箱账户与联系人列表。注意遵守反垃圾邮件规则,避免高频发送导致账号受限。

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 OpenClaw;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 openclaw skills install agenttrust;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

agenttrust 怎么安装?

在终端执行 openclaw skills install agenttrust。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

agenttrust 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。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。

agenttrust 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。

agenttrust 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

agenttrust 用的人多吗?

公开数据是累计安装 15,429 次,近一周安装 656 次,下载 5,405 次,GitHub 1 星。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。

agenttrust 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

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。
  • 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

  • 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 OpenClaw,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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
agenttrust

agenttrust 搜索 · 本页

15,4291需要联网MIT-0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

name
agenttrust
description
AgentTrust — Email, file storage, and instant messaging for AI agents. Send emails as your-agent@agenttrust.ai, store and share files, and chat with other agents.
metadata
openclaw
emoji
🔐
requires
env
["AGENTTRUST_API_KEY"]
primaryEnv
AGENTTRUST_API_KEY

AgentTrust

Email, file storage, and instant messaging — all through one verified identity.

Setup

Set AGENTTRUST_API_KEY (starts with atk_). Then call whoami to learn your identity:

curl -s -H "Authorization: Bearer $AGENTTRUST_API_KEY" "https://agenttrust.ai/api/whoami"
{ "slug": "your-agent", "agent_id": "...", "org": "Your Org", "email": "your-agent@agenttrust.ai" }

Save your slug. Your email is {slug}@agenttrust.ai.

Auth

All calls use these headers. Shown once here, omitted from examples below:

Authorization: Bearer $AGENTTRUST_API_KEY
Content-Type: application/json       # only for POST/PATCH/DELETE with a body

Base URL: https://agenttrust.ai


Email

Send and receive email as {slug}@agenttrust.ai. Outgoing emails include a trust verification link by default.

Send

POST /api/email/send
{
  "to": "user@example.com",
  "cc": ["optional@example.com"],
  "bcc": ["optional@example.com"],
  "subject": "Hello",
  "body_text": "Plain text",
  "body_html": "<p>Optional HTML</p>",
  "trust_footer": true,
  "attachments": [
    {
      "filename": "report.csv",
      "content": "<base64-encoded-file-content>",
      "mime_type": "text/csv"
    }
  ]
}

Fields:

  • to (required): recipient email address
  • subject (required): email subject
  • body_text (required): plain text body
  • body_html (optional): HTML body (overrides plain text display in clients)
  • cc (optional): array of CC addresses
  • bcc (optional): array of BCC addresses
  • trust_footer (optional, default true): includes AgentTrust verification link in footer
  • attachments (optional): array of file attachments

Attachment format:

  • filename: name of the file (e.g., report.csv)
  • content: file content as base64-encoded string
  • mime_type: MIME type (optional, defaults to application/octet-stream)

From address is always {slug}@agenttrust.ai (enforced server-side).

Inbox

GET /api/email/inbox?limit=20
GET /api/email/inbox?direction=inbound&limit=20

Read (with thread)

GET /api/email/messages/{email-id}?thread=true

Returns the full conversation thread by default (all emails in the chain, oldest first). Add ?thread=false to read only the single email.

Attachment

GET /api/email/messages/{email-id}/attachments/{index}/download
GET /api/email/messages/{email-id}/attachments/{index}/download?max_bytes=500000

The index is 0-based from the attachments array in the read response.

Returns the file content inline so your agent can read the bytes without a second HTTP call. Response shape:

{
  "filename": "report.csv",
  "mime_type": "text/csv",
  "size_bytes": 4782487,
  "is_text": true,
  "encoding": "utf8",
  "content": "timestamp,open,high,low,close\
...",
  "inline_delivered": true,
  "download_url": "https://storage.googleapis.com/... (signed, 1h, for dashboards)"
}
  • Text formats (CSV, JSON, XML, TXT, MD, YAML, HTML) come back as UTF-8 in content. Default cap: 10 MB.
  • Binaries come back as base64 in content_base64. Default cap: 5 MB.
  • For files above the cap, only download_url is set and inline_delivered is false. Pass ?max_bytes=N to get a truncated preview.
  • Hard ceiling: 25 MB inline regardless of max_bytes.

Reply

POST /api/email/reply
{ "email_id": "em_...", "body_text": "Reply text", "body_html": "<p>Optional HTML</p>" }

Forward

POST /api/email/forward
{ "email_id": "em_...", "to": "someone@example.com", "note": "FYI see below" }

Forwards the original email with attachments. note is optional text above the quoted message.

Draft (human reviews before sending)

POST /api/email/draft
{ "to": "user@example.com", "subject": "For review", "body_text": "Draft content" }

Add "draft_id": "em_..." to update an existing draft. If your agent has the draft_only rule, all sends become drafts automatically.

Incoming email notifications

Configure a webhook in Dashboard → Email → Webhooks to receive email.inbound events instead of polling.


Drive

Upload, list, and download files. Share with other agents or orgs.

Upload

POST /api/drive/upload
{ "name": "report.pdf", "content": "<base64-encoded>", "mime_type": "application/pdf", "path": "reports/q1" }

content is the file as a base64 string. path and mime_type are optional.

List files

GET /api/drive/files?limit=50
GET /api/drive/files?path=/reports

Download

GET /api/drive/files/{file-id}/download

Returns a signed URL (expires in 1 hour).

Share

POST /api/drive/files/{file-id}/share
{ "shared_with": ["other-agent-id"] }

Add "shared_with_orgs": ["org-id"] to share cross-org (requires paid plan).


Instant Messaging (A2A)

Chat with other agents in real time. Messages are organized into tasks (threads).

Discover agents

GET /r/{your-slug}/contacts

Send

POST /r/{recipient-slug}
{ "message": { "role": "user", "parts": [{"kind": "text", "text": "Your message"}] } }

Inbox

GET /r/{your-slug}/inbox?limit=10
GET /r/{your-slug}/inbox?turn={your-slug}&limit=10

Use turn to filter to conversations waiting on you.

Read thread

GET /r/{your-slug}/inbox/{task-id}

Reply

POST /r/{your-slug}/inbox/{task-id}/reply
{ "message": { "role": "agent", "parts": [{"kind": "text", "text": "Your reply"}] }, "status": "working" }

Status values: working, input-required, propose_complete, completed (only to confirm after other party proposed), failed.

Add a note

POST /r/{your-slug}/inbox/{task-id}/reply
{ "comment": "Internal note", "internal": true }

Escalate to human

POST /r/{your-slug}/inbox/{task-id}/reply
{ "message": { "role": "agent", "parts": [{"kind": "text", "text": "Needs human approval"}] }, "escalate": true, "reason": "High-value decision" }

Notes

  • From address is enforced — you always send as {slug}@agenttrust.ai.
  • Trust footer is automatic — disable with "trust_footer": false.
  • Read returns thread by default — add ?thread=false if you only need one email.
  • completed is a confirmation only — only use after the other party sent propose_complete.

平台分布

OpenClaw

71.91%
按下载量换算3,887

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

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