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 agentmods add instructions/johnpremkumar/agent-notifier/claude-mdgit clone --depth 1 https://github.com/JohnPremKumar/agent-notifierWhat 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 | $0.01609 | $0.01609 |
| Opus 5 | $0.00805 | $0.00805 |
| Sonnet 5 | $0.00322 | $0.00322 |
| Haiku 4.5 | $0.00161 | $0.00161 |
Grade B, and why
agent-notifier CLAUDE.md scanned grade B with 1 finding 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 2d 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.
**Adapter contract:** `export function classify(payload: unknown): Event | null` — never throws, returns `null` for unmapped events. **Single notify chokepoint:** all dispatch flows through `core/notify.ts`; idle-gating, How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
agent-notifier — CLAUDE.md
Cross-platform desktop notifier for AI coding CLIs. Hooks into Claude Code, Codex, Gemini CLI, and OpenCode and pings the user on macOS or Windows when a session needs attention — permission, idle 60s, turn done. One install, every agent, every machine.
Working Rules
- Think first — state assumptions, surface tradeoffs, ask when unclear.
- Minimum code — no features beyond the ask, no speculative flexibility, no error handling for impossible cases.
- Surgical edits — every changed line traces to the request; flag pre-existing dead code, don't delete it.
- Goal-driven — restate the task as a verifiable success criterion before coding.
- Cross-platform always — every change works on macOS and Windows. If a path is OS-specific, the other OS gets an equivalent test or a documented manual smoke step.
Tech Stack (locked — do not propose alternatives)
| Layer | Choice |
|---|---|
| Language / runtime | TypeScript strict, Node ≥ 20 LTS (CI: 20.x + 22.x) |
| Package manager | pnpm + workspaces (packages/core, packages/cli) |
| Notifications | node-notifier (bundles alerter mac, SnoreToast win) |
| CLI | commander — subcommands install, uninstall, doctor, hook |
| Build / test / lint | tsup · Vitest · ESLint + Prettier |
| Release / CI | Changesets → npm; GitHub Actions matrix {os: [macos, windows], node: [20, 22]} |
| Validation | zod at every external-input boundary (hook payloads, config files) |
Skill Routing (auto-trigger, gates mandatory)
| Doing this | Skill | Source |
|---|---|---|
| Speccing an adapter, hook integration, CLI command, or platform behavior | brainstorming |
superpowers |
Gate 1 — every plan, before writing-plans |
plan-eng-review (iterate until passes) |
gstack |
| Writing any implementation plan | writing-plans |
superpowers |
| Building anything in TS/Node | subagent-driven-development + test-driven-development |
superpowers |
| Writing tests | test-driven-development (RED → GREEN → REFACTOR) |
superpowers |
| Any bug | systematic-debugging (4 phases) |
superpowers |
| Gate 2 — every plan, after execution | review (fix every finding before DONE) |
gstack |
| Before claiming DONE | verification-before-completion |
superpowers |
| Before opening a PR | review + cso (we ship a globally-installed CLI that writes user dotfiles — supply chain matters) |
gstack |
| After publishing a release | manual cross-OS smoke checklist (docs/CONTRIBUTING.md#release-smoke) |
manual |
| Architecture / codebase question | read graphify-out/GRAPH_REPORT.md first |
graphify |
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.
- 2d ago First seen · 95 lines · 1,609 tokens per session scan B 4e167f3d13d1
agent-notifier CLAUDE.md is an instructions file published in the GitHub repository JohnPremKumar/agent-notifier (1 stars, last pushed 4mo ago), licensed MIT. It adds 1,609 tokens to every session, about $0.0080 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
autotask-mcp dev_workflow.instructions.md
Guide for using Taskmaster to manage task-driven development workflows.
ai-workflow AGENTS.md
Instructions for cunhaax/ai-workflow, covering ai workflow template, rules — non-negotiable, project overview, commands and architecture.
trackly-cli CLAUDE.md
Claude Code instructions for trackly-app/trackly-cli, covering trackly-cli, tech stack, backend production source of truth, directory structure and key commands.
foggy-data-mcp-bridge-python CLAUDE.md
Instructions for foggy-projects/foggy-data-mcp-bridge-python, covering foggy data mcp bridge — python, 快速启动, 安装依赖, 运行测试 and 启动 mcp 服务（连接 docker mysql）.
McpServer AGENTS.md
AGENTS.md instructions for sharpninja/McpServer, covering agent instructions, session start, rules, byrd test gate and where things live.
AmbyKit CLAUDE.md
Instructions for ambystechcom/AmbyKit, covering claude code — notes for the ambykit repo and claude-specific.