Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.wednesdayai.dev/llms.txt

Use this file to discover all available pages before exploring further.

​
Messaging basics

Talk to WednesdayAI the same way you would with any contact — just send a message. There is no special syntax for normal conversation.

​
Starting a conversation

Send any message to the bot on your connected channel (WhatsApp, Telegram, etc.). The agent responds to your message, using your conversation history as context. First time? See Pairing your first channel if you have not yet been connected to the gateway.

​
How context works

WednesdayAI keeps a session — a running window of conversation history — that provides context for each response. The agent knows what you said earlier in the session. Sessions do not persist forever. When a session ends (inactivity timeout, or you start a new one), the agent starts fresh. To start a fresh conversation: 
/reset
Or equivalently: 
/new
You can also start fresh with a different model: 
/new claude-3-5-sonnet

​
Getting status

Send /status at any time to see:
  • Which AI model is active
  • Gateway connectivity
  • Provider usage and quota (if tracking is enabled)
  • Active sessions

​
Setting context in your message

You can steer the conversation with natural language — no special commands needed. To be explicit, use directives:
  • /think high — ask the model to reason carefully before responding
  • /verbose on — show detailed tool output and reasoning steps
  • /model <name> — switch to a specific model for this session
These can be combined with a message: 
/think high
Explain how database indexes work and when to use them.

​
Long conversations

If your conversation is getting long and the agent is losing context or making mistakes, use: 
/compact
This summarizes the conversation so far and continues with a fresh context window. You can optionally add instructions: 
/compact Focus on the authentication module we discussed.

​
Tips

  • Tool use is automatic — if the agent needs to search the web or run code, it will. Just ask naturally.
  • Multiple messages — the agent processes one message at a time. If you send several quickly, they are queued.
  • Inline shortcuts — some commands work embedded in a message. /help Where are the docs? gives a help reply and then processes your question.
  • Voice — if your admin has enabled TTS, use /tts always to receive audio responses.

​
Quick command reference

CommandEffect
/reset or /newStart a fresh conversation
/statusCurrent model, gateway status, quota
/helpShow available commands
/think <off/low/high>Adjust model reasoning depth
/model <name>Switch AI model
/compactSummarise context and continue
/verbose onShow tool output and reasoning
/usage tokensShow token count on each reply
Full command reference: Slash commands.