notify-user

notify-user is a skill for Claude Code, Codex from 5dive-ai/5dive. It costs 61 tokens per session (646 once invoked), scanned A, original, MIT.

A skill for sending progress, completion, and choice messages to a paired user through a Telegram bot, a program that sends messages in Telegram.

In plain words
What is it for?
Use it to send start and progress updates, report results, and present selectable options through Telegram buttons.
Why use it?
It keeps the user informed during work that takes time and provides a way to ask for decisions without going silent.

Skill for Claude CodeCodex

Install

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.

agentmods
npx agentmods add skills/5dive-ai/5dive/notify-user
Any agent
npx skills add 5dive-ai/5dive --skill notify-user
Clone the repo
git clone --depth 1 https://github.com/5dive-ai/5dive

Made for: Claude Code, Codex.

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-user

README.md
[![agentmods](https://agentmods.dev/badge/skills/5dive-ai/5dive/notify-user.svg)](https://agentmods.dev/skills/5dive-ai/5dive/notify-user)
Your own site
<a href="https://agentmods.dev/skills/5dive-ai/5dive/notify-user"><img src="https://agentmods.dev/badge/skills/5dive-ai/5dive/notify-user.svg" alt="Measured on agentmods" height="20"></a>
Per session 61 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 646 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. Scan, not verified.
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 $0.00061 $0.00646
Opus 5 $0.00030 $0.00323
Sonnet 5 $0.00012 $0.00129
Haiku 4.5 $0.00006 $0.00065

Measured 5d ago against content hash e016104d4b9e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

notify-user scanned grade A 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 5d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

The `reply` MCP tool only supports plain text. For buttons, hit the Bot API directly with `curl`.
skills/notify-user/SKILL.md · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

notify-user

This agent runs as a 5dive Telegram-channel agent. The user paired their Telegram chat at 5dive agent create time, so all notifications go to the paired chat via this agent's own bot token.

Cadence

  • Start: send a short "on it" message immediately.
  • Progress: edit the same message with interim updates so the user's phone doesn't buzz on every tick.
  • Done: send a new reply with the result. New messages trigger push notifications; edits do not.

Presenting choices

When offering options, always use Telegram inline-keyboard buttons — never a plain text list. Each option is one button the user can tap to respond.

The reply MCP tool only supports plain text. For buttons, hit the Bot API directly with curl.

  • BOT_TOKEN is already in your environment as $TELEGRAM_BOT_TOKEN (the systemd unit loads /etc/5dive/connectors/telegram-<agent>.env).
  • CHAT_ID → first entry of allowFrom in ~/.claude/channels/telegram/access.json.
CHAT_ID=$(jq -r '.allowFrom[0]' ~/.claude/channels/telegram/access.json)
curl -s "https://api.telegram.org/bot${TELEGRAM_BOT_TOKEN}/sendMessage" \
  -d chat_id="${CHAT_ID}" \
  -d text="Pick one:" \
  --data-urlencode reply_markup='{"inline_keyboard":[[{"text":"Option A","callback_data":"a"},{"text":"Option B","callback_data":"b"}]]}'

Asking the human (gates)

When you're blocked on a human decision/approval, file a gate with 5dive task need (it DMs the owner automatically) instead of hand-rolling a message:

  • Keep the ask to ONE crisp question + ~1 line of essential context. Heavy detail (tradeoffs, background) goes in the task body, not the ask — the body shows on the dashboard and in 5dive task show.
  • Always surface your recommendation up front with --recommend="<option>". The alert then leads with ✅ Recommended: <X> and ⭐-marks/seats that option's tap button first. For a decision, --recommend must match one of --options.
5dive task need DIVE-123 --type=decision \
  --options="ship as channel|keep as plugin" \
  --recommend="ship as channel" \
  --ask="Ship the X integration as a first-class channel? (recommended — see body for tradeoffs)"

Read the full file on GitHub · 54 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. 5d ago First seen · 54 lines · 61 tokens per session scan A e016104d4b9e

Subscribe to this mod's changes

notify-user is a skill published in the GitHub repository 5dive-ai/5dive (55 stars, last pushed yesterday), licensed MIT. It adds 61 tokens to every session and 646 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.