@@ -24,7 +24,7 @@ Works with VS Code family of editors (Cursor, Antigravity, VSCodium, etc.).
 
 **LLMs are brilliant, yet mindless pattern matchers.** Because they lack properties similar to mental models, they can't _imagine_ how the correct output should look like _before_ their context is filled with files.
 
-Coding agents address this limitation via tool calling—although keyword search may work, it requires instructions complex enough to fully understand your intent. For complex codebases, the process may run for dozens of minutes and still yield mixed results.
+Coding agents address this limitation via so-called tool calling. The model starts each task from a clean slate, deciding by itselft what files are necessary based on the provided instructions. This puts pressure on your prompting skills and effort, also on model itself because file-collecting self-dialogue is a complexity added to the primary task.
 
 **CWC proposes an alternative, more mindful 🧠 approach through context engineering.** Pick files and be certain what the model gets to see. Type simpler instructions and get edits in seconds with accuracy that can't get any better!