Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add castle-x/skills-x --skill feishu-notifygit clone --depth 1 https://github.com/castle-x/skills-xWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/castle-x/skills-x/feishu-notify)<a href="https://agentmods.dev/skills/castle-x/skills-x/feishu-notify"><img src="https://agentmods.dev/badge/skills/castle-x/skills-x/feishu-notify/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/castle-x/skills-x/feishu-notify"><img src="https://agentmods.dev/badge/skills/castle-x/skills-x/feishu-notify.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00140 | $0.02611 |
| Opus 5 | $0.00070 | $0.01306 |
| Sonnet 5 | $0.00028 | $0.00522 |
| Haiku 4.5 | $0.00014 | $0.00261 |
Grade B, and why
feishu-notify scanned grade B with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 10d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Reads agent configuration directoriesmediumAgent snooping
.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.
配置文件路径:`~/.claude/settings.json`(全局生效,对所有项目有效)。 Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -s -X POST "$WEBHOOK_URL" \ How it starts
The opening of the file, as written. The whole thing — 277 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Feishu Notify — Claude Code 飞书通知 Skill
每当 Claude Code 完成任务(触发 Stop Hook)时,自动向飞书群发送一条消息卡片通知,内容包含会话 ID、完成时间、停止原因、最后使用的工具、工作目录、主机名、系统运行时间,以及 Claude 最后一条回复的摘要(前 300 字符)。
执行流程
Step 1:引导用户获取飞书群机器人 Webhook 地址
首先询问用户是否已有飞书群机器人 Webhook 地址。如果没有,引导用户按以下步骤操作:
1. 打开飞书,进入任意群组(或新建一个专用的「Claude 通知」群)
2. 点击右上角「群设置」图标 → 「群机器人」→ 「添加机器人」
3. 选择「自定义机器人」
4. 填写机器人名称(如:Claude Code 助手)
5. 点击「添加」,复制生成的 Webhook 地址
格式示例:https://open.feishu.cn/open-apis/bot/v2/hook/xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx
获取到 Webhook 地址后,继续执行 Step 2。
Step 2:创建目录并写入通知脚本
创建 ~/.claude/ 目录,并将以下脚本写入 ~/.claude/notify-feishu.sh。
将脚本中的 __WEBHOOK_URL__ 替换为用户提供的真实 Webhook 地址。
#!/usr/bin/env bash
# 飞书通知脚本 - Claude Code Stop Hook
# 保存路径:~/.claude/notify-feishu.sh
WEBHOOK_URL="__WEBHOOK_URL__"
# 从 stdin 读取 Claude Code Hook 传入的 JSON
INPUT=$(cat)
# 提取 Hook 字段
SESSION_ID=$(echo "$INPUT" | jq -r '.session_id // "unknown"')
STOP_HOOK_REASON=$(echo "$INPUT" | jq -r '.stop_hook_reason // empty')
TOOL_NAME=$(echo "$INPUT" | jq -r '.tool_name // empty')
TRANSCRIPT_PATH=$(echo "$INPUT" | jq -r '.transcript_path // empty')
# 提取环境信息
HOSTNAME=$(hostname 2>/dev/null || echo "unknown")
WORK_DIR=$(pwd)
TIMESTAMP=$(date '+%Y-%m-%d %H:%M:%S')
UPTIME=$(uptime -p 2>/dev/null || uptime | sed 's/.*up/up/')
SHORT_SESSION="${SESSION_ID:0:12}"
# 提取最后一条 assistant 文本回复(截取前 300 字符)
LAST_REPLY=""
if [[ -n "$TRANSCRIPT_PATH" && -f "$TRANSCRIPT_PATH" ]]; then
LAST_REPLY=$(grep '"type":"assistant"' "$TRANSCRIPT_PATH" \
| while IFS= read -r line; do
echo "$line" | jq -r '
select(.message.content[]?.type == "text")
| [.message.content[] | select(.type == "text") | .text]
| join("")' 2>/dev/null
done \
| tail -1 \
| cut -c1-300)
ORIGINAL_LEN=$(grep '"type":"assistant"' "$TRANSCRIPT_PATH" \
| while IFS= read -r line; do
echo "$line" | jq -r '
select(.message.content[]?.type == "text")
| [.message.content[] | select(.type == "text") | .text]
| join("")' 2>/dev/null
done | tail -1 | wc -c)
[[ "$ORIGINAL_LEN" -gt 300 ]] && LAST_REPLY="${LAST_REPLY}..."
fi
# ⚠️ 关键:用数组 + printf '%s\n' 产生真实换行符
# 不能用 "\n" 字符串拼接,否则飞书会渲染为字面 \n 而不是换行
LINES=()
LINES+=("📋 **会话 ID**:\`${SHORT_SESSION}...\`")
LINES+=("⏰ **完成时间**:${TIMESTAMP}")
[[ -n "$STOP_HOOK_REASON" ]] && LINES+=("📝 **停止原因**:${STOP_HOOK_REASON}")
[[ -n "$TOOL_NAME" ]] && LINES+=("🔧 **最后工具**:${TOOL_NAME}")
LINES+=("📂 **工作目录**:\`${WORK_DIR}\`")
LINES+=("🖥️ **主机名称**:${HOSTNAME}")
LINES+=("🔌 **系统运行**:${UPTIME}")
if [[ -n "$LAST_REPLY" ]]; then
LINES+=("")
LINES+=("💬 **最后回复**:")
LINES+=("${LAST_REPLY}")
fi
# printf '%s\n' 将数组每个元素转为真实换行的多行字符串
MD_CONTENT=$(printf '%s\n' "${LINES[@]}")
# ⚠️ 关键:用 jq -n --arg 传入多行字符串,jq 会正确转义为合法 JSON
# 不能用 echo/printf 手动拼接 JSON,否则换行符会破坏 JSON 结构
CARD_JSON=$(jq -n \
--arg md "$MD_CONTENT" \
'{
msg_type: "interactive",
card: {
schema: "2.0",
config: { wide_screen_mode: true },
header: {
template: "green",
title: { tag: "plain_text", content: "✅ Claude Code 任务完成" }
},
body: {
elements: [{ tag: "markdown", content: $md }]
}
}
}')
# 发送通知,静默失败不影响 Claude Code 正常运行
curl -s -X POST "$WEBHOOK_URL" \
-H 'Content-Type: application/json' \
-d "$CARD_JSON" \
> /dev/null 2>&1 || true
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 10d ago First seen · 277 lines · 140 tokens per session scan B a3a450eaab52
feishu-notify is a skill published in the GitHub repository castle-x/skills-x (18 stars, last pushed 4mo ago), licensed MIT. It adds 140 tokens to every session and 2,611 once invoked, about $0.0007 per session on Opus 5. A static security scan graded it B with 2 findings (reads agent configuration directories, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
orbit-notion
Open Orbit briefing skill — selected by the Orbit pipeline when Notion is the user's only connected connector, or when the user explicitly scopes their daily digest to Notion. Pulls the past 24 hours of document edits, comments, mentions, and database row changes from the user's authenticated Notion connection and…
Cortex
Operate Cortex, the LifeOS memory system — the typed Knowledge Archive (People, Companies, Ideas, Research with typed related: links) plus recall of prior work sessions, ISAs, and conversations. Search, add, harvest, develop, ingest, distill, graph-navigate, recall. USE WHEN cortex, knowledge, knowledge base, search…
pinchtab-mcp
Use this skill when a task requires browser automation through PinchTab's MCP server connected to a remote browser instance. Covers navigation, element interaction, data extraction, form filling, multi-step flows, and session management via MCP tools.
feishu
Work with Feishu or Lark bots, docs, sheets, bitables, approval flows, and OpenAPI/MCP setup without hardcoding credentials.
peekaboo
Capture and automate macOS UI with the Peekaboo CLI.
mochi-remind
Handle due reminders — notify the user with natural language and mark them done.