Google Ads Clicker is a module inside Botlord Platform. Install and activate the platform first, then configure this engine.
Advanced options
Debug mode, AI steering, cloud sync, and keeping the platform current.
3 min readAI steering modes and providers
There are two distinct AI modes — Assisted and Autonomous. They are different features and are enabled differently.
AI Assisted mode
AI Assisted mode keeps the bot on its normal route (search → SERP → click → dwell) but uses an LLM to analyse the page before clicking. The AI reads page composition — ad density, AI Overview blocks, competing results — and recommends which result to click and when. The bot still executes the click mechanically.
Enable AI Assisted in the AI tab. No prompt required. It adds a small cost per session (one LLM call per SERP) but improves click quality, especially on competitive or dynamic keywords.
AI Autonomous mode (AI Agent tasks)
AI Autonomous mode replaces the fixed click route with a fully AI-directed session. Select AI Agent as the task type in the campaign builder, then write a natural-language prompt. The AI plans every step from scratch — what to search, which result to click, how long to stay, which links to follow.
Example prompt: 'You are a marketing manager evaluating project management tools. Search Google for the best project management software for agencies, read a couple of organic results, then find a sponsored ad and click it. Browse the landing page as if you are genuinely considering signing up.'
The AI produces a unique session each time based on the prompt. Use this mode when you need varied, realistic engagement patterns that are difficult to detect as scripted. It consumes more API tokens and runs slower than standard tasks.
AI providers
OpenAI GPT-4o is the default and most reliable for both Assisted and Autonomous modes. Anthropic Claude handles complex multi-step prompts particularly well — a good choice for AI Agent tasks that require nuanced navigation.
OpenRouter routes to dozens of models from a single API key, which is useful for cost control or testing different models. DeepSeek and Google Gemini are lower-cost alternatives for simpler use cases.
Local servers (Ollama, LM Studio) work offline with any OpenAI-compatible model. Smaller local models often struggle with multi-step navigation prompts — use a 7B+ model and test before running unattended.
Set your API key and select your provider in the AI tab. Run a single manual session with Logs open before using AI features in unattended production runs.
Cloud campaign sync
Campaigns you build in the web campaign builder at thebotlord.com are synced to the desktop when the bot starts. Cloud campaigns take priority over local config on conflict.
If the sync fails (no internet, server down), the bot falls back to the last-saved local config. It never hard-fails because of a cloud sync error.
Use Save to Cloud in the campaign builder to push local campaigns back to the cloud for backup or multi-machine sharing.
Debug mode and logs
Enable debug in Configuration to get verbose per-step logging. Disable it for production runs — it adds overhead and fills logs quickly.
The Logs tab shows each session step: search, consent, scroll, click, dwell, exit. Filter by worker ID when running multiple concurrent workers.
Screenshots can be triggered on error when that option is enabled. Useful for diagnosing exactly where a session failed.
Updates
Use Check for Updates from the About section. The platform auto-updates on the stable channel. Beta builds are a separate channel — only install if you are actively testing pre-release features.
After any major update, run a short manual session before leaving anything unattended on a VPS.