# Telegram Reaction Prober

> Probe which emoji reactions are accepted in a specific Telegram chat/message, record an allow/deny list, and optionally remove test reactions afterwards.

- Skill: `kbarbel640-del/telegram-reaction-prober` (Agent Skill, multi-file: 4 files)
- Install (CLI): `npx skillmds@latest add kbarbel640-del/telegram-reaction-prober`
- Raw SKILL.md: https://api.skillmd.com/api/skills/kbarbel640-del/telegram-reaction-prober/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: kbarbel640-del (https://skillmd.com/u/kbarbel640-del)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/kbarbel640-del/telegram-reaction-prober

---


# Telegram Reaction Prober

## What this skill is
Telegram reactions are **chat-specific**: a bot can only react with emojis that are enabled for that chat/message. When you try an unsupported emoji, Telegram returns:

- `400 Bad Request: REACTION_INVALID`

This skill is the *prober* (the method + starter datasets), not a global list.
It helps you:
- test a list of candidate emojis against a specific `message_id`
- classify them as **Allowed** vs **Rejected**
- optionally remove successful reactions after testing (so you don't spam the chat)
- write the results into `TOOLS.md` (or another file)

> Important: **Do not publish your private chat’s allow/deny list** as “the answer for everyone”.
> Share the prober + candidate sets; each user still probes their own chat.

## Limits / Reality check
There is no practical way to "test every emoji" (Unicode is enormous, and this Clawdbot Telegram integration does not expose Telegram's enabled-reaction list directly).

So the best approach is:
1) Test a **curated emoji set** you care about (common reactions)
2) Save the whitelist for that chat
3) Re-run when Telegram chat settings change

## How to run (manual)
1) Pick a target Telegram `message_id` (e.g. the most recent message in the chat).
2) Pick a candidate emoji set:
   - start small (15–30), or
   - use the included 200-emoji starter list:
     - `skills/telegram-reaction-prober/assets/emoji200-unicode-frequency-2019.txt`
3) For each emoji:
   - call `message` tool with `action=react` and that emoji
   - if it succeeds → mark Allowed
   - if it fails with `REACTION_INVALID` → mark Rejected
4) Cleanup options:
   - **Fast/quiet:** don’t remove; Telegram generally keeps only one reaction per user/bot (so you’re effectively just flipping it)
   - **Clean:** remove successful reactions after testing via `remove=true`
5) Write results to `/home/ubuntu/clawd/TOOLS.md` under a heading like:
   - `### Telegram reactions (this chat) — tested on message_id XYZ`

## Suggested starter emoji set
Allowed in many chats:
- 👍 ❤️ 🔥 🙏 🎉 🤔 👀

Often rejected (depends on chat):
- ✅ 😂 💡

Good candidates to test next:
- 👎 😅 😭 😮 😍 🤝 👏 🙌 💯 😡 😴 🧠 🧩 ⚠️

## Notes
- Be mindful of rate limits; don’t blast 100+ reactions without delays.
- Keep tests on a single known `message_id` and record it.