#context-window
Every post tagged "context-window".
36 posts · showing 21–30
-
Why Turn 20 Costs More Than Turn 2
A six-turn run whose final context is 12,000 tokens sends 42,000. That gap is agent economics — and it makes an early oversized tool result the most expensive thing in your run.
-
When Memory Makes an Agent Worse
Every retrieved memory is content injected into the context, relevant or not. Three ways that degrades the agent — including the one where it argues with a user about their own stated preference.
-
Few-Shot Examples: When They Help an Agent and When They Hurt
Two or three similar examples is the worst case — enough to teach a template, not enough to convey the range. And the example nobody includes is the one where the right answer is to decline.
-
The File System as a Tool: Underrated Agent Infrastructure
Four hundred thousand rows won't fit in a context window; a path and a preview cost nothing. A scoped directory solves large results, durable state, and reviewable artifacts at once — if you resolve paths correctly.
-
Give the Agent a Scratchpad: External Working Memory
Three small tools stop the context from being the only place findings can live. Compaction stops being lossy, long runs stop growing, and you get a readable record instead of a forty-turn transcript.
-
What a System Prompt Should and Shouldn't Contain
Five things belong in a system prompt. Most of what accumulates there — tool instructions, deterministic rules, incident patches, data — belongs somewhere it would actually work.
-
Chunking for Agents Is Not Chunking for Search
Search results are read by a person who fills in the gaps. An agent doesn't — it acts on the fragment. That one difference changes how you chunk, how many you return, and what "no results" should mean.
-
Run, Session, or Global? Scoping Agent Memory Correctly
Too-small a scope produces a visible annoyance; too-large produces a silent wrong answer. One question assigns any fact correctly, and one rule keeps multi-tenant reads from becoming an incident.
-
Format Tool Output for a Model, Not for a Log File
A bare empty array means "none exist" and "we couldn't check" at the same time, and the agent will pick the confident reading. Six changes to what your handlers return, in order of payoff.
-
Where You Put Things in the Context Changes the Answer
Same instructions, same documents, different results — because the constraint sat in the middle of a long array. Two ordering rules, one of which also decides whether your prompt cache ever hits.