Getting it into your agent
There is no command for this one: it runs only inside a plugin, and the catalogue could not identify which plugin ships it. The source is linked below.
Wrote 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/instructions/mariozzj/cc-notify-hooks/claude-md)<a href="https://agentmods.dev/instructions/mariozzj/cc-notify-hooks/claude-md"><img src="https://agentmods.dev/badge/instructions/mariozzj/cc-notify-hooks/claude-md.svg" alt="Measured on agentmods" 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 | $0.02573 | $0.02573 |
| Opus 5 | $0.01287 | $0.01287 |
| Sonnet 5 | $0.00515 | $0.00515 |
| Haiku 4.5 | $0.00257 | $0.00257 |
Grade B, and why
cc-notify-hooks CLAUDE.md 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 6d 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.
- 启用 hooks 需要 `~/.codex/config.toml` 添加 `[features]\ncodex_hooks = true` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **依赖**: `jq`（JSON 解析）、`curl`（HTTP 请求）、`osascript`（macOS 通知） How it starts
The opening of the file, as written. The whole thing — 151 lines — stays where its author put it; the contents beside it link to each section on GitHub.
cc-notify-hooks
Claude Code、Codex CLI、Reasonix 与 dsh（DeepSeek Harness）的分级推送通知系统。通过 hook/拦截点触发多渠道通知，用户响应后自动取消排队中的推送。
技术栈
- 语言: Bash（核心脚本）+ JavaScript（dsh 宿主插件，零依赖）
- 依赖:
jq（JSON 解析）、curl（HTTP 请求）、osascript（macOS 通知） - 集成: Claude Code / Codex CLI / Reasonix 的 hooks 插件机制 + dsh 的 cordis 拦截点
项目结构
plugins/cc-notify-hooks/ # 真实插件根目录（根目录的 scripts/hooks/config 是软链）
├── scripts/notify.sh # 主调度器：事件过滤、延迟排队、渠道分发（兼容 4 个 agent 字段）
├── scripts/clear_pending.sh# 清除待发通知（用户交互时触发）
├── scripts/pre_tool_use.sh # 提问工具 dispatcher（request_user_input / ask / AskUserQuestion）
├── scripts/channels/*.sh # 11 个渠道实现
├── hooks/hooks.json # Claude Code hook 事件定义
├── hooks/codex-hooks.json # Codex CLI hook 事件定义
├── reasonix-plugin.json # Reasonix 原生插件清单（v2，payloadFormat=claude）
├── dsh-plugin/ # dsh 宿主插件（cordis，订阅 5 个拦截点）
├── .claude-plugin/ # Claude Code 插件清单
├── .codex-plugin/ # Codex CLI 插件清单
├── config/notify.example.json
└── skills/config/SKILL.md # 交互式配置 skill（Claude Code 专属）
install.sh # 独立安装入口（路由）
install/{claude,codex,reasonix,dsh}.sh
test_notify.sh # 渠道连通性 + 模板/agent 识别测试
常用命令
# 安装
bash install.sh # 交互式选择 Claude Code / Codex / Reasonix / dsh
bash install.sh claude|codex|reasonix|dsh
# 测试
bash test_notify.sh # 测试所有已启用渠道
bash test_notify.sh bark # 测试单个渠道
bash test_notify.sh hook # 模拟 Claude Code hook 流程
bash test_notify.sh codex # 模拟 Codex CLI PermissionRequest 事件
bash test_notify.sh agents # 验证 Reasonix / dsh agent 识别
node plugins/cc-notify-hooks/dsh-plugin/test/plugin.test.mjs # dsh 插件单元测试
# 插件模式运行
claude --plugin-dir ./cc-notify-hooks
reasonix plugin install ./plugins/cc-notify-hooks --link --yes
核心机制
- 分级延迟: 短通知（秒级：macOS/Bark/Telegram）→ 长通知（分钟级：微信/飞书/钉钉/Slack）
- Pending 取消: 发送前创建标记文件，用户响应时清除，后台进程检查标记决定是否发送
- 速率限制: 同类事件默认 10 秒内不重复推送
- 事件过滤: 跳过子 agent、Stop 循环保护、/exit 静默
配置
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.
- 6d ago First seen · 151 lines · 2,573 tokens per session scan B be14a9c244f5
cc-notify-hooks CLAUDE.md is an instructions file published in the GitHub repository MarioZZJ/cc-notify-hooks (19 stars, last pushed 18d ago), licensed MIT. It adds 2,573 tokens to every session, about $0.0129 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 instructions, from other repositories
lattice CLAUDE.md
Claude Code instructions for techygarg/lattice: Read and follow PROJECT.md in this repository. All Lattice conventions, structure, skill rules, patterns, and anti-patterns live there. This file exists so Claude Code loads project context from the repo root.
forge CLAUDE.md
Claude Code instructions for TT-Wang/forge, covering forge, quick start, skills, agents and mcp tools.
builder-essential-skills AGENTS.md
Instructions for tamdogood/builder-essential-skills, covering agent notes and cinematic banner typography.
claude-proj-blueprint CLAUDE.md
Claude Code instructions for maduwanthasathsara0-hub/claude-proj-blueprint, covering project, tech stack, architecture, documentation directories and code conventions.
dex AGENTS.md
AGENTS.md instructions for exmergo/dex, covering dex: driving the engine from any agent, the loop: explore, transform, maintain (etm), the command contract, or, with the pinned wrapper a skill ships and the envelope.
virtual-company CLAUDE.md
Instructions for k1lgor/virtual-company, covering 🏢 virtual company: claude code plugin, 🛑 the iron law, 📢 completion status protocol, 🧠 operational self-improvement (learning log) and 📜 core operating principles.