notify-me-setup

notify-me-setup is a skill for Claude Code from jovesun-lab/whetstone. It costs 106 tokens per session (1,394 once invoked), scanned A, original, MIT.

A record-keeping tool for measuring an AI agent's performance in each session and comparing results over time. It stores a simple Markdown trend table, with an optional script for charts.

In plain words
What is it for?
Use it to benchmark, score, grade, or compare an agent across sessions, including A/B tests between two setups. It tracks results over a longer run rather than judging one session alone.
Why use it?
It can be difficult to tell whether a model, instruction setup, or skill change actually helped or caused a regression. Repeated measurements make those changes easier to compare.

Skill for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: names the AskUserQuestion tool.

Runs only inside its plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else. Install the plugin, not this.

Part of the notify-me plugin — 1 skill, 3 hooks shipped together

Good fit Use it to benchmark, score, grade, or compare an agent across sessions, including A/B tests between two setups. It tracks results over a longer run rather than judging one session alone.

Compare 6 skills from other repositories ↓
Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code
/plugin marketplace add jovesun-lab/whetstone
Claude Code
/plugin install notify-me

Made for: Claude Code.

Or install notify-me, the plugin that ships this one along with the rest of its 1 skill, 3 hooks.

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.

agentmods badge for notify-me-setup

README.md
[![agentmods](https://agentmods.dev/badge/skills/jovesun-lab/whetstone/notify-me-setup.svg)](https://agentmods.dev/skills/jovesun-lab/whetstone/notify-me-setup)
Your own site
<a href="https://agentmods.dev/skills/jovesun-lab/whetstone/notify-me-setup"><img src="https://agentmods.dev/badge/skills/jovesun-lab/whetstone/notify-me-setup.svg" alt="Measured on agentmods" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,394 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00106 $0.01394
Opus 5 $0.00053 $0.00697
Sonnet 5 $0.00021 $0.00279
Haiku 4.5 $0.00011 $0.00139

Measured 7d ago against content hash 560dcdbd1c6f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

notify-me-setup scanned grade A with 0 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 7d 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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

notify-me/skills/notify-me-setup/SKILL.md · 86 lines

How it starts

The opening of the file, as written. The whole thing — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.

notify-me setup

This skill configures the notify-me plugin by writing ~/.notify-me/config, which the notification hook reads. Run the whole flow in plain chat. Do NOT use a popup/AskUserQuestion — present every choice as lettered text options in the message and wait for a typed reply.

Config file format

Write ~/.notify-me/config as sourceable shell KEY="value" lines:

TOKEN="user-chosen-secret"   # this IS the ntfy "topic" you subscribe to on your phone
SOUND="on"
SOUND_FILE="/System/Library/Sounds/Glass.aiff"
NOTIFY_ON_STOP="off"
SERVER="https://ntfy.sh"

Field meaning:

  • TOKEN — the private secret the phone subscribes to. In the ntfy app this value is called the "topic". Acts as a password, so it must be random/unguessable.
  • SOUNDon/off. Whether to play a local sound on this computer.
  • SOUND_FILE — absolute path to a sound file, OR the literal @system to play the OS's currently-selected alert sound (macOS "beep" / Windows beep). A real file path is more reliable than @system, because @system follows the separate OS "alert volume" which is easy to leave at zero.
  • NOTIFY_ON_STOPon/off. If on, also alert when the agent finishes a turn (fires often — default off).
  • SERVER — ntfy server base URL. Leave as https://ntfy.sh unless the user self-hosts.

First-run / setup flow

If ~/.notify-me/config already exists, read it and show current settings first, then ask which field to change. Otherwise run full onboarding:

Step 1 — pick the alert sound list (from THIS machine)

Run the bundled scanner to list sounds actually present on the user's computer:

bash "${CLAUDE_PLUGIN_ROOT}/scripts/list-sounds.sh"

Present the returned files as a lettered list (label by filename, e.g. "A) Glass B) Ping C) Submarine ..."). Offer to preview any sound before choosing — on macOS preview with afplay "<path>". Let the user pick one, or pick a sensible default (Glass on macOS) if they don't care — prefer a real file over @system for reliability. Always offer @system as one option for users who'd rather have the alert follow whatever sound they pick in their OS sound settings (warn that it depends on the OS "alert volume"). If the scanner returns nothing (e.g. Windows, or an unusual setup), offer @system, ask for an absolute path to a sound file, or set SOUND="off" (phone push still works).

Read the full file on GitHub · 86 lines

Changes

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.

  1. 7d ago First seen · 86 lines · 106 tokens per session scan A 560dcdbd1c6f

Subscribe to this mod's changes

notify-me-setup is a skill published in the GitHub repository jovesun-lab/whetstone (8 stars, last pushed 17d ago), licensed MIT. It adds 106 tokens to every session and 1,394 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories