Skip to main content
Bot platform adapters now ship in the praisonai-bot package. praisonai bot serve still works exactly as documented here; for a standalone install see praisonai-bot Migration.
Status reactions update an emoji on the user’s message to reflect run state — queued, thinking, tool execution, done, or error — and add a “still working…” ⏳ / “taking longer than expected” ⚠️ signal when a run goes quiet, all without reading reply text. 
from praisonaiagents import Agent
from praisonai.bots import TelegramBot

agent = Agent(name="assistant", instructions="Helpful assistant")
bot = TelegramBot(token="...", agent=agent, status_reactions=True)
bot.start()
The user sends a message on Telegram; emoji reactions update through queued, thinking, tool, done, or error states.

​
Quick Start

1

Enable with one flag

from praisonaiagents import Agent
from praisonai.bots import TelegramBot

agent = Agent(name="assistant", instructions="Helpful assistant")
bot = TelegramBot(token="...", agent=agent, status_reactions=True)
bot.start()
2

Customise emoji via dict

bot = TelegramBot(
    token="...",
    agent=agent,
    status_reactions={
        "thinking_emoji": "💭",
        "tool_emoji": "⚙️",
        "done_emoji": "🎉",
        "debounce_delay": 1.0,
    },
)
3

Full control with StatusConfig

from praisonai.bots import TelegramBot, StatusConfig

bot = TelegramBot(
    token="...",
    agent=agent,
    status_reactions=StatusConfig(
        thinking_emoji="💭",
        immediate_terminal=True,
    ),
)

​
How It Works

The user sends a message and watches a single emoji on that message move through each run stage.

​
Stall Signals

When the agent falls quiet for too long — a slow LLM or a stuck tool — the same reaction slot switches to ⏳ after stall_soft_s (default 20 s) and ⚠️ after stall_hard_s (default 60 s), and any real progress clears the signal immediately. 
from praisonaiagents import Agent
from praisonai.bots import TelegramBot

agent = Agent(name="assistant", instructions="Helpful assistant")
bot = TelegramBot(
    token="...",
    agent=agent,
    status_reactions={
        "stall_soft_s": 15,   # ⏳ after 15s of silence
        "stall_hard_s": 45,   # ⚠️ after 45s
    },
)
bot.start()

​
Label Fallback (channels without reactions)

On channels where capabilities["reactions"] is False — WhatsApp, Email, and Slack in “label” mode — the same phase state machine renders as a single status label the bot edits in place (queued → thinking… → using a tool… → done) instead of an emoji reaction, and the adapter picks the mode automatically from the channel’s capabilities with no extra config.

​
Configuration Options

OptionTypeDefaultDescription
queued_emojistr"⏳"Shown when the run is queued
thinking_emojistr"🤔"Shown while the LLM is generating
tool_emojistr"🔧"Shown during tool execution
done_emojistr"✅"Terminal: success
error_emojistr"❌"Terminal: failure
debounce_delayfloat0.5Seconds before applying intermediate states
immediate_terminalboolTrueApply done/error immediately
stall_soft_sfloat20.0Seconds without progress before ⏳ “still working…”
stall_hard_sfloat60.0Seconds without progress before ⚠️ “taking longer than expected”
enabledboolFalseMaster switch — off by default; the bot adapter opts in when you pass status_reactions=True
Reactions auto-skip on channels where capabilities["reactions"] is False (WhatsApp, Email). On Slack, Unicode emoji map to Slack :text_name: form (e.g. 🤔 → thinking_face).
Telegram’s reaction API is set-based — remove_reaction clears all bot reactions on the message rather than a specific emoji. This is a Telegram API limitation.

​
Best Practices

Avoids hitting reaction rate limits on busy Telegram groups.
Helps users tell when an external API call is in progress.
Users see done/error states instantly without debounce delay.
If a legitimate tool takes 40 s (web scraping, image generation), raise stall_soft_s to 45–60 s so users don’t see ⏳ while everything is still healthy.
Reactions silently degrade to a status label edit — no per-channel code required.
See Run Status Controller for the low-level API.

Channel Capabilities

Which channels support reactions

Typing Indicators

Complementary typing feedback

Streaming Replies

Live token-by-token responses