Auto_Mark (Streamer.bot C# Implementation)

This repo contains two copy/paste-ready C# scripts for Streamer.bot's Execute C# Code sub-action:

• Listener_ChatLogger.cs (chat logger / short-term memory)
• Brain_PerpetualResponse.cs (AI response generator)

Default personality and naming are now set for Auto_Mark (resident AI mod / robotic co-host persona).

1) Listener Action (Chat Logger)

1. In Streamer.bot, create Action: Auto_Mark - Listener.
2. Add sub-action: Execute C# Code.
3. Paste code from Listener_ChatLogger.cs.
4. Add Trigger: Twitch > Chat Message.

What it does:

• Reads Global Var chat_buffer.
• Appends each incoming message as [User]: Message.
• Trims buffer to ~1000 chars.
• Saves back to Global Var chat_buffer.
• Skips self-logging when the sender matches perpetual_bot_name.

2) Brain Action (AI Reply)

1. Create Action: Auto_Mark - Brain.
2. Add sub-action: Execute C# Code.
3. Paste code from Brain_PerpetualResponse.cs.
4. Add Trigger:
   • Command Trigger: e.g. !auto_mark (recommended), OR
   • Chat Message Trigger with your chosen keyword logic.

What it does:

• Pulls chat_buffer, current user name, and current message.
• Pulls User Var perpetual_lore for that user.
• Builds the model prompt using configurable bot name + system prompt.
• Uses ai_provider to pick Gemini or OpenAI key + default endpoint/model.
• Parses and posts model response to chat.
• Logs API errors via CPH.LogInfo.

3) Settings You Can Change in Streamer.bot

Set these in Global Variables:

• perpetual_bot_name = default Auto_Mark
• ai_provider = gemini or openai (default gemini)
• gemini_api_key = your Gemini key
• openai_api_key = your OpenAI key
• ai_endpoint = optional override (provider default used if empty)
• ai_model = optional override (provider default used if empty)
• perpetual_system_prompt = full Auto_Mark persona prompt
• chat_buffer = empty string initially

Backward compatibility:

• Shared key variable ai_api_key is still supported.
• Legacy google_api_key is checked for Gemini if gemini_api_key is missing.

Optional per-user variable:

• perpetual_lore (User Variable), e.g. Fumbles every platform jump.

4) Default Auto_Mark Persona Included

The default system prompt implements this behavior:

• resident AI mod and robotic co-host in TheDeutschMark universe
• self-aware bot created/coded by Mark
• witty, meta-humor, sarcastic, but useful for moderation/help
• recurring Botzandra obsession lore joke
• references to channel identity (TheDeutschMark, Jacob & Willie) when relevant
• concise, streamer-safe replies

5) Publish/Share Extension Checklist

1. Create both actions and paste scripts.
2. Create triggers:
   • Auto_Mark - Listener -> Twitch Chat Message
   • Auto_Mark - Brain -> !auto_mark command
3. Pre-create global variables listed above with defaults.
4. Export as a Streamer.bot extension package.
5. In your shared README/changelog include:
   • required Streamer.bot version
   • required Twitch account auth state
   • variable table from section 3
   • first-run instructions to set ai_provider and provider key
6. Test import on a clean Streamer.bot profile before publishing.

6) What You Need To Do Right Now

1. Paste both scripts into their corresponding Streamer.bot actions.
2. Set ai_provider (gemini or openai) and matching API key variable.
3. Confirm globals use defaults (or customize name/model/prompt).
4. Send test chat messages to populate chat_buffer.
5. Run !auto_mark roast me.