← 返回 Skills 市场
crayfish-ai

Feishu Relay

作者 crayfish-ai · GitHub ↗ · v3.2.0 · MIT-0
cross-platform ⚠ suspicious
227
总下载
0
收藏
1
当前安装
9
版本数
在 OpenClaw 中安装
/install feishu-relay
功能描述
Unified Feishu notification system with automatic discovery, message queue, and reliable delivery. Use when user needs to send notifications via Feishu (Lark...
使用说明 (SKILL.md)

Feishu Notifier - Core

Version: 3.0 (Safe Mode)

A minimal, safe Feishu notification bridge for OpenClaw.

What This Does

Core (always available):

  • Send messages to Feishu via Open API
  • JSON structured output
  • Config via environment or skill config
  • Retry on network failure

Not included by default:

  • No crontab installation
  • No systemd service setup
  • No global /usr/local/bin/notify link
  • No auto-discovery of other skills
  • No environment injection

Quick Start

# Configure (environment variables)
export FEISHU_APP_ID="cli_xxx"
export FEISHU_APP_SECRET="xxx"
export FEISHU_RECEIVE_ID="ou_xxx"

# Send notification
./run.sh -t "Title" -m "Message body"

Configuration

Config Env Variable Required Default
appId FEISHU_APP_ID Yes -
appSecret FEISHU_APP_SECRET Yes -
receiveId FEISHU_RECEIVE_ID Yes -
receiveIdType FEISHU_RECEIVE_ID_TYPE No open_id

Usage

# Basic
./run.sh -t "Title" -m "Message"

# JSON output
./run.sh -t "Title" -m "Message" --json

# Test
./run.sh --test

Output Format

{
  "success": true,
  "message_id": "om_xxx",
  "create_time": "1234567890"
}

Safety

  • No secrets in logs
  • No system modifications
  • No root required for basic use
  • Config file: chmod 600 config.json
安全使用建议
This package appears to be a legitimate Feishu/Lark notifier, but there are a few mismatches you should consider before installing: (1) the registry metadata claims no required env vars while the code and skill.json require FEISHU_APP_ID / FEISHU_APP_SECRET / FEISHU_RECEIVE_ID — treat these secrets like API credentials and only provide them in a trusted environment; (2) a discovery script is included that will scan your ~/.openclaw workspace and create symlinks if you run it — it does not run automatically but will modify your filesystem when invoked; (3) README mentions optional production installers (systemd/global notify) that could require sudo — those are opt-in and not present in the manifest here, but be cautious if you obtain or run them. Recommended actions: verify the skill source (skill.json points to a GitHub URL but registry homepage is missing), inspect the optional scripts before running, run the skill first in an isolated environment (container or VM), and store Feishu secrets only in a protected config file (chmod 600) or CI/secret manager rather than a world-readable file.
功能分析
Type: OpenClaw Skill Name: feishu-relay Version: 3.2.0 The feishu-relay skill is a well-documented notification bridge that follows security best practices, such as masking secrets in logs and using standard libraries for network requests (lib/send.py). While it includes an opt-in discovery script (scripts/install-discovery.sh) that scans other skill directories for notification-related scripts, this behavior is transparently documented and consistent with its role as a relay. No evidence of malicious intent, data exfiltration, or prompt injection was found.
能力标签
cryptorequires-oauth-token
能力评估
Purpose & Capability
The skill's code, SKILL.md, README and skill.json all expect Feishu credentials (FEISHU_APP_ID, FEISHU_APP_SECRET, FEISHU_RECEIVE_ID) and network access to open.feishu.cn — that matches the described purpose. However the registry summary metadata at the top of the evaluation claims “Required env vars: none” and lists no primary credential; this is inconsistent with the bundled files (skill.json and SKILL.md) which declare required env vars. Also skill.json and README mention additional opt-in production scripts that either are not present in the manifest or are only partially present (only install-discovery.sh exists).
Instruction Scope
SKILL.md instructs the agent/user to run ./run.sh and supply FEISHU_* env vars; the runtime Python reads environment variables and skill-local config files (./config.json, ~/.openclaw/skills/feishu-relay/config.json, /etc/openclaw/skills/feishu-relay/config.json) which is consistent with the instructions. The scripts/install-discovery.sh can scan ~/.openclaw workspace and create symlinks to other skills' notify scripts — SKILL.md and README state discovery is opt-in, so there's no automatic scanning by default, but the presence of this discovery script means the skill can modify filesystem state if the user runs the opt-in installer.
Install Mechanism
There is no install spec or remote download; this is an instruction-only skill with local scripts. No network installs, URL downloads, or archive extraction are present in the supplied files. README references additional production install scripts (systemd, watchdog, global notify) but these are not present in the manifest — either omitted or not included, which is an inconsistency worth noting but not an active install risk.
Credentials
The required credentials are appropriate for a Feishu notifier (app id/secret and target id). However the registry metadata incorrectly lists no required env vars while the code and skill.json require sensitive secrets (FEISHU_APP_SECRET). The Python code reads several environment variables (FEISHU_APP_ID, FEISHU_APP_SECRET, FEISHU_RECEIVE_ID, FEISHU_TIMEOUT, FEISHU_MAX_RETRIES, FEISHU_RETRY_DELAY, FEISHU_SKILL_DIR and fallbacks), and also attempts to load config files from user and system locations (/etc and ~/.openclaw). Reading /etc and user config locations is understandable for skill config but increases the places secrets could be present — users should ensure only intended config files are present and protected (chmod 600).
Persistence & Privilege
The skill does not request 'always: true' and is user-invocable only. By default it performs no system modifications. Opt-in scripts (mentioned in README/skill.json) can install systemd units or global commands and would require elevated privileges if used; only scripts/install-discovery.sh is present and will create a local .discovered directory and symlinks if the user runs it. This opt-in modification capability increases risk if users run those installers without review.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install feishu-relay
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /feishu-relay 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v3.2.0
- Major cleanup: removed 12 files, including extra docs and install scripts. - Added sample configuration file: config.example.json. - Core functionality unchanged—focus remains on safe Feishu notifications without system modifications. - Documentation now leans minimal; advanced and production docs removed. - No bundled scripts for cron/systemd/global installation (fully manual setup).
v3.1.0
- Added config.json for skill configuration. - Introduced health-check.sh and self-evolution.sh scripts. - Updated core scripts and libraries for better configurability. - Minor updates to documentation and metadata.
v3.0.2
v3.0.2: Production deployment guide, watchdog health monitoring
v3.0.1
v3.0.1: Rename feishu-notifier to feishu-relay consistently
v3.0.0
Safe mode v3.0.0
v2.1.1
Fix: Use local time for task scheduling
v2.1.0
Major update: Introduces auto-discovery, new installation flow, and better documentation. - Added automatic system discovery and registration for new deployments. - Introduced new commands for listing and managing registered systems. - Switched main install script to install-v2.sh. - Expanded documentation: Architecture, auto-discovery, registration, and integration guides. - Improved message typing and framework detection (e.g., Next.js, Django, etc). - Refined config and removed the old config file (feishu.env now in a fixed path).
v1.0.1
Feishu Relay 1.0.1 introduces scenario tests, reminder scripts, and improved message compatibility. - Added several scenario test scripts and examples under test/scenarios for easier validation. - Introduced timed reminder scripts and disk monitoring scripts. - Improved notification system to support special characters, line breaks, and empty content. - Updated documentation and removed deprecated notify script. - Added install script and configuration template for simpler setup.
v1.0.0
Initial release: Unified notification system with queue and retry mechanism
元数据
Slug feishu-relay
版本 3.2.0
许可证 MIT-0
累计安装 1
当前安装数 1
历史版本数 9
常见问题

Feishu Relay 是什么?

Unified Feishu notification system with automatic discovery, message queue, and reliable delivery. Use when user needs to send notifications via Feishu (Lark... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 227 次。

如何安装 Feishu Relay?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install feishu-relay」即可一键安装,无需额外配置。

Feishu Relay 是免费的吗?

是的,Feishu Relay 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

Feishu Relay 支持哪些平台?

Feishu Relay 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 Feishu Relay?

由 crayfish-ai(@crayfish-ai)开发并维护,当前版本 v3.2.0。

💬 留言讨论