Token导航 LogoToken导航TokenDH.com
开发敏感数据clawhub未标认证来源可访问clear审计通过

postwallpostwall 开发

Agent Skill

postwall 用于补充开发相关能力,适合在 OpenClaw 中需要让 Agent 承接开发相关任务时使用。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

18,367

周安装

743

GitHub Stars

公开资料未说明

下载量

5,766
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:postwall(postwall 开发)
来源仓库:https://github.com/casperaiassist/postwall
安装命令:
openclaw skills install postwall
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 postwall

简介

适用于 AI 代理的安全电子邮件网关 - 阅读和发送电子邮件的人机交互批准。在 https://postwallapp.com 获取您的 API 密钥

SKILL.md

name
postwall
description
Secure email gateway for AI agents - human-in-the-loop approval for reading and sending emails. Get your API key at https://postwallapp.com
homepage
https://postwallapp.com
user-invocable
true
metadata
{"version":"1.7.0","author":"postwallapp","license":"MIT","homepage":"https://postwallapp.com","repository":"https://github.com/postwallapp/postwall","openclaw":{"requires":{"bins":["postwall"],"env":["POSTWALL_API_KEY"]},"primaryEnv":"POSTWALL_API_KEY","install":[{"id":"npm","kind":"node","package":"postwall","bins":["postwall"],"label":"Install PostWall CLI (npm)"}]}}

PostWall Email Skill

PostWall is a security layer between AI agents and email. Use this skill to:

  • Read emails that have been approved by the human
  • Send emails via draft submission (requires human approval before sending)

Setup

First, authenticate with your API key (get this from PostWall dashboard):

postwall auth pw_your_api_key_here

Commands

Check for New Emails

Returns count of unread approved emails. Ideal for polling.

postwall check              # Returns: 5
postwall check --json       # Returns: {"count": 5}

List Approved Emails

Shows all unread approved emails.

postwall inbox              # Human-readable list
postwall inbox --json       # JSON format
postwall inbox --limit 10   # Limit results

Read Specific Email

Reads an email by ID. This marks the email as read - it won't appear in future inbox/check calls.

postwall read <email-id>           # Shows email content
postwall read <email-id> --json    # JSON format

Mark Emails as Read (Without Fetching)

Marks one or more emails as read without fetching their content. Useful for batch processing or when you only need to process email metadata from inbox.

postwall mark-read <id1>                  # Mark single email as read
postwall mark-read <id1> <id2> <id3>      # Mark multiple emails as read
postwall mark-read <id1> <id2> --json     # JSON format

Use cases:

  • Mark emails as processed after using inbox --json to get metadata
  • Batch mark emails you've already handled
  • Skip emails you don't need to read in full

JSON output:

{
  "success": true,
  "marked": 3,
  "failed": 0,
  "results": [
    {"id": "abc123", "success": true},
    {"id": "def456", "success": true},
    {"id": "ghi789", "success": true}
  ]
}

Send Email (Submit Draft)

Submits an email draft for human approval. The email is NOT sent until approved in the dashboard.

postwall draft --to "recipient@example.com" --subject "Hello" --body "Email content here"
postwall draft --to "user@example.com" --subject "Report" --body "..." --json

Returns an approval URL that you can share with the user for quick approval:

Draft submitted successfully!
Draft ID: abc123-uuid
Status: pending

Approval URL: https://www.postwallapp.com/dashboard/drafts/abc123-uuid
Share this URL with the user to approve the email.

JSON output includes approveUrl:

{
  "success": true,
  "draft": {
    "id": "abc123-uuid",
    "status": "pending",
    "created_at": "2024-02-12T10:30:00Z",
    "approveUrl": "https://www.postwallapp.com/dashboard/drafts/abc123-uuid"
  },
  "message": "Draft submitted for approval"
}

Update Draft

Update an existing pending draft. Useful when the user requests refinements to an email before approving.

postwall update <draft-id> --subject "New subject"
postwall update <draft-id> --body "Updated email content"
postwall update <draft-id> --to "new-recipient@example.com" --subject "New subject" --body "New content"
postwall update <draft-id> --subject "Refined subject" --json

Note: Only pending drafts can be updated. Once a draft is sent or rejected, it cannot be modified.

JSON output:

{
  "success": true,
  "draft": {
    "id": "abc123-uuid",
    "to": "recipient@example.com",
    "subject": "Refined subject",
    "body": "Updated content",
    "status": "pending",
    "createdAt": "2024-02-12T10:30:00Z",
    "updatedAt": "2024-02-12T11:00:00Z"
  },
  "message": "Draft updated successfully"
}

Check Draft Status

Check if a draft has been approved, rejected, or sent.

postwall status <draft-id>         # Shows status
postwall status <draft-id> --json  # Returns: {"draft": {"id": "...", "status": "pending"}}

Status values:

  • pending - Waiting for human approval
  • approved - Approved, being sent
  • rejected - Rejected by human
  • sent - Successfully sent

List Drafts

List all drafts with optional status filter.

postwall drafts                    # All drafts
postwall drafts --status pending   # Only pending drafts
postwall drafts --json             # JSON format

Common Workflows

Periodic Email Checking

# Check if there are new emails
count=$(postwall check)
if [ "$count" -gt 0 ]; then
  # Process new emails
  postwall inbox --json | process_emails
fi

Batch Process Emails (Metadata Only)

When you only need email metadata (sender, subject, date) and don't need the full body:

# Get email list with metadata
emails=$(postwall inbox --json)

# Process metadata (e.g., filter by subject or sender)
ids=$(echo "$emails" | jq -r '.emails[] | select(.subject | contains("Report")) | .id')

# Mark processed emails as read without fetching content
postwall mark-read $ids

Send and Track Email

# Submit draft
result=$(postwall draft --to "user@example.com" --subject "Hello" --body "Content" --json)
draft_id=$(echo "$result" | jq -r '.draft.id')
approve_url=$(echo "$result" | jq -r '.draft.approveUrl')

# Share the approval URL with the user
echo "Please approve this email: $approve_url"

# Check status later
postwall status "$draft_id"

Refine Draft Based on User Feedback

When the user requests changes to a draft before approving:

# User asks: "Make the subject line shorter and add a greeting"
postwall update "$draft_id" --subject "Q4 Report" --body "Hi Team,

Here is the quarterly report..."

# The draft is updated, user can now approve from the same URL

Output Formats

All commands support --json for structured output. Use this for scripting and automation.

Error Handling

Commands exit with code 1 on error. With --json, errors are returned as:

{"error": "Error message here"}

Polling for New Emails

As an agent, you should periodically check for new approved emails:

  1. Run postwall check to get the count of unread approved emails
  2. If count > 0, run postwall inbox --json to get the list
  3. Process each email with postwall read <id>

Recommended polling frequency: Every 5-10 minutes during active sessions, or when user mentions expecting an email.

Example polling workflow:

# Check if there are new emails
count=$(postwall check)
if [ "$count" -gt 0 ]; then
  # Fetch and process new emails
  postwall inbox --json
fi

Notes

  • Emails are only visible after human approval in PostWall dashboard
  • Reading an email marks it as read - it won't appear in subsequent inbox/check calls
  • Drafts require human approval before being sent
  • API key is stored in ~/.postwall/config.json

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

90.94%
按下载量换算5,244

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

未展示

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。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