List Telegram Dialogs
List all chats, groups, and channels visible to your personal Telegram account.
Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.
Preflight
- Session must exist:
~/.local/share/telethon/<profile>.session- If missing, run
/tlg:setupfirst
- If missing, run
Usage
/usr/bin/env bash << 'DIALOGS_EOF'
SCRIPT="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/tlg}/scripts/tg-cli.py"
# Default profile
uv run --python 3.13 "$SCRIPT" dialogs
# Specific profile
uv run --python 3.13 "$SCRIPT" -p missterryli dialogs
# Filter results
uv run --python 3.13 "$SCRIPT" dialogs | grep -i "search term"
DIALOGS_EOF
Additional Commands
Read Messages
read returns the full text of every message by default. Multi-line
bodies are indented under the header so the message stays visually
grouped. No truncation.
# Full text (default — recommended)
uv run --python 3.13 "$SCRIPT" read <chat_id> -n 10
# Short scan listing — truncate each body to N chars (\n flattened to "⏎")
uv run --python 3.13 "$SCRIPT" read <chat_id> -n 50 --preview 200
Use --preview N only when you're scanning many messages and want a
single-line summary per row. For routine reading, omit it — long messages
deserve to be read in full, not silently cut at 200 chars (the prior default,
which forced repeated direct-Telethon escape hatches when content mattered).
Account Info
uv run --python 3.13 "$SCRIPT" whoami
Output Format
Chat Name (id: 1234567890)
Use the id value with send-message skill to send to that chat.
Post-Execution Reflection
After this skill completes, check before closing:
- Did the command succeed? — If not, fix the instruction or error table that caused the failure.
- Did parameters or output change? — If tg-cli.py's interface drifted, update Usage examples and Parameters table to match.
- Was a workaround needed? — If you had to improvise (different flags, extra steps), update this SKILL.md so the next invocation doesn't need the same workaround.
Only update if the issue is real and reproducible — not speculative.