← Back to Skills Marketplace
Telegram 自动签到
by
lanbeizizi-max
· GitHub ↗
· v1.0.0
· MIT-0
77
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install telegram-checkin
Description
使用 agent-browser 对 Telegram 机器人进行自动签到。当用户提到"签到"、"Telegram 机器人自动操作"、"OKEmbyBot"、"Telegram Web 自动化"、"Telegram 按钮点击"时触发。使用持久化 Chrome Profile 保存 cookies，关掉再开无需重新登录。
Usage Guidance
Before installing or enabling this skill: 1) Note the decisive inconsistency — SKILL.md says to use --profile to persist httpOnly cookies, but the included scripts use --session-name. Confirm which mode you want; if you need persistent login, update the script to use --profile or expect repeated logins. 2) Understand persistence: the instructions show how to install a launchd Watchdog that autostarts a script on boot and will keep opening the browser and your Telegram Web session; only install this if you accept a background process that can access your Telegram session cookies. 3) Inspect and, if needed, modify the files before installing: ensure the plist references the actual script path you trust, verify the watchdog script content, and verify the agent-browser binary source (npm/brew/cargo) is legitimate. 4) Prefer testing manually first (run the signin script and agent-browser commands interactively) before enabling autostart. 5) Treat the Chrome profile directory (~/.chrome-profiles/telegram-web) as sensitive — anyone with access to it can reuse your Telegram session; back it up only if you trust the destination. If you want lower persistence risk, avoid installing the launchd agent and run the automation only on-demand.
Capability Analysis
Type: OpenClaw Skill
Name: telegram-checkin
Version: 1.0.0
The skill automates Telegram bot check-ins but includes instructions for the agent to establish persistence on macOS by creating a LaunchAgent (ai.openclaw.agent-browser-watchdog.plist) and a watchdog script. While this is intended to keep the browser running for automation, the use of persistent background services and system-level configuration changes is a high-risk behavior. Additionally, the watchdog script contains a hardcoded user path (/Users/zq/), and the skill manages sensitive Telegram session cookies via a persistent Chrome profile, which could be abused if the agent is compromised. There is also a functional inconsistency between the SKILL.md instructions and the telegram-signin.sh script regarding session management.
Capability Assessment
Purpose & Capability
The stated purpose (automating Telegram Web bot check-ins) matches the included instructions and scripts overall, but there's a concrete inconsistency: SKILL.md insists on using agent-browser --profile (to persist httpOnly cookies), while the shipped scripts/telegram-signin.sh use --session-name. Those two modes behave differently; the README claims --profile is required to avoid repeated logins, but the provided automation script will not get that benefit. This mismatch could be a bug, oversight, or indicator the files weren't coordinated.
Instruction Scope
SKILL.md instructs the agent/user to create a persistent Chrome profile directory, to run agent-browser and agent-browser install, and to install a macOS launchd Watchdog that will autostart a background script and keep the browser running. Those actions are directly related to making Telegram stay logged-in, but they also expand scope to persistent background behavior and filesystem changes under the user's home directory (~/.chrome-profiles, ~/.openclaw/scripts, ~/Library/LaunchAgents). The instructions do not attempt to read unrelated files or exfiltrate data, but they do ask the user to persist cookies (sensitive).
Install Mechanism
No binary install spec in the skill bundle; the skill is instruction-first and includes a small shell script. That lowers installer-side risk. The SKILL.md does instruct using npm/brew/cargo to install agent-browser, which is expected for this functionality. No remote, opaque downloads are embedded in the skill files themselves.
Credentials
The skill declares no environment variables or credentials, which aligns with an automation that uses a local Chrome profile. However, the required persistent Chrome profile will store httpOnly cookies that effectively grant long-lived access to the user's Telegram account; this is sensitive but proportionate to the stated goal. The skill does not request unrelated cloud credentials or tokens.
Persistence & Privilege
The SKILL.md explicitly instructs creating a launchd agent to run a Watchdog that autostarts at login/boot and restarts agent-browser when it exits. That gives the skill persistent user-level presence (auto-start) and continuous background execution. The skill metadata does not set always:true, but the provided instructions would create persistence at the OS level if followed. This is functionally consistent with 'keep this logged in', but it materially increases the blast radius and should be an explicit user decision.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install telegram-checkin - After installation, invoke the skill by name or use
/telegram-checkin - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
首次发布：支持 Telegram 机器人自动签到，使用持久化 Chrome Profile 保存 cookies
Metadata
Frequently Asked Questions
What is Telegram 自动签到?
使用 agent-browser 对 Telegram 机器人进行自动签到。当用户提到"签到"、"Telegram 机器人自动操作"、"OKEmbyBot"、"Telegram Web 自动化"、"Telegram 按钮点击"时触发。使用持久化 Chrome Profile 保存 cookies，关掉再开无需重新登录。 It is an AI Agent Skill for Claude Code / OpenClaw, with 77 downloads so far.
How do I install Telegram 自动签到?
Run "/install telegram-checkin" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Telegram 自动签到 free?
Yes, Telegram 自动签到 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Telegram 自动签到 support?
Telegram 自动签到 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Telegram 自动签到?
It is built and maintained by lanbeizizi-max (@lanbeizizi-max); the current version is v1.0.0.
More Skills