Don't Show AI What It Doesn't Need
100 tools all in system? Token explosion. This calls for on-demand loading design
THE QUESTION THIS PAGE ANSWERS
ANSWER FIRSTWhat is the key idea behind “Don't Show AI What It Doesn't Need”?
100 tools all in system? Token explosion. This calls for on-demand loading design
Make the claim earn its place. Use this page as a decision aid, not a definition to memorize. Connect the idea to one real task, one observable result, and one failure that would change your mind.
Write one question you could answer with evidence after trying this idea.
A conclusion that sounds complete but leaves the key assumption untested.
Three steps:
① First turn: names only
The System Prompt only includes tool names + a one-line summary. The AI just needs to know the capability exists.
② Expand on demand
When the AI decides to call a tool, the system dynamically injects the full description and parameter schema.
③ Retract after use
After the tool is used, the next turn no longer includes the full description — back to names only.
How “Full Load vs. Lazy Load” changes an answer
“① First turn: names only The System Prompt only includes tool names + a one-line summary.” shows that a model does not process the “word count” we see. It processes Token pieces. Tokenization affects input length, how much context fits, and how much computation a request consumes.
Length, information, and context are different
As “① First turn: names only The System Prompt only includes tool names + a one-line summary.” grows, separate three questions: how many Tokens the text becomes, which pieces can change the current decision, and whether older material has fallen outside the context window. Removing repetition is often more useful than simply making the window larger.
Keep what can change the decision
Use “① First turn: names only The System Prompt only includes tool names + a one-line summary.” as an A/B test: keep the same question while removing repeated background, compressing format, and trimming irrelevant history. Compare answer quality, latency, and Token count.
From “Full Load vs. Lazy Load” to “Drag to See the Difference”
“Full Load vs. Lazy Load” grounds the problem in “Full Load ~34,000 tk Lazy Load ~2,800 tk”. “Drag to See the Difference” then moves it toward “Adjust tool count to see how Token cost changes Number of tools: 50 17,500 Full Load Tokens 1,550 Lazy Load Tokens 91% Savings”. Together, they show that the lesson is not just a conclusion to remember, but a claim with conditions.
Carry the judgment into the next situation
For long text, keep what can change the conclusion before compressing format and history. A larger context is worth its cost only when the added information is useful.
- “Full Load vs. Lazy Load”: Full Load ~34,000 tk Lazy Load ~2,800 tk
- “Drag to See the Difference”: Adjust tool count to see how Token cost changes Number of tools: 50 17,500 Full Load Tokens 1,550 Lazy Load Tokens 91% Savings
- “The closing point”: ① First turn: names only The System Prompt only includes tool names + a one-line summary. The AI just needs to know the capability exists. ② Expand on demand When the AI decides to call a tool, the system dynam…
The final “The closing point” brings the discussion to “① First turn: names only The System Prompt only includes tool names + a one-line summary. The AI just needs to know the capability exists. ② Expand on demand When the AI decides to call a tool, the system dynam…”. The useful thing to carry forward is knowing which judgments must be revisited when input, scale, or risk changes.
I turned one judgment from this article into a small experiment I could run today. Knowing what to observe next is more useful than simply remembering the conclusion.
After reading this, I first looked for the conditions behind the idea instead of copying the method into a project. That order made the later trade-offs much clearer.
When this judgment reaches real work, which constraint should be added first? I am curious which step matters most between reading and the first practical attempt.
No discussion on this article yet.