Hidden Gems & Power User Tips
Unlock the full potential of Boring with these advanced features and lesser-known tricks.
🛠️ Deep Debugging with Profiles
You might know about lite vs full profiles, but did you know you can customize them?
Why boring_debug_code Vanished?
If you are in lite mode (default), advanced debugging tools are hidden to save context. To get them back:
Custom Profiles
Create custom_profile.toml to mix-and-match:
[profiles.my_debug]
include = ["boring_read_file", "boring_debug_code", "boring_security_scan"]
exclude = ["boring_commit"]
🧠 Brain Surgery
Your AI learns from mistakes (~/.boring/brain), but sometimes it learns the wrong thing.
View Your Brain
Check what patterns are currently stored:
Force Unlearning
If Boring keeps making a mistake it "learned" to do, you can manually edit this file or use the boring_learn tool explicitly:
🕵️ Mastering the Router
The boring() router uses semantic scoring, not just keywords.
Triggering Analysis
To force a "Sequential Thinking" session, use keywords like:
- "Think about..."
- "Analyze deeply..."
- "Reason through..."
External Knowledge (Context7)
You don't need to leave your IDE to check documentation.
Pro Tip: Be specific with library names. "Query PyTorch docs" works better than "Query ML docs".
🛡️ Shadow Mode Tricks
Strict Mode for Code Review
You can use Shadow Mode as a forced code review tool, even for safe operations.
Now, every single file write will pop up a diff for you to approve. It's like an interactive PR review.
🔍 HyDE Search Secrets
Boring uses HyDE (Hypothetical Document Embeddings) for search.
- Bad Query: "auth error"
- Good Query: "Why does the login function return a 401 error when the token is valid?"
Why? HyDE generates a fake "perfect answer" and searches for code that looks like that. Phrasing as a question helps it hallucinate a better search target.