nafyi.

Improved token efficiency for longer agent runs · Cursor

Cursor Blog ·

The authors describe token-efficiency work in Cursor’s agent harness. They report a 7% reduction in user token costs without reducing agent quality, a roughly 66% reduction in the system prompt, a 46.9% reduction in total tokens for sessions calling an MCP tool after dynamic tool loading, and a 20% reduction in cold cache misses after cache-related changes. Lies 4 Standpunkte mit Belegen und Links zu den Originalquellen.

Jediah Katz, Connor O’Keefe & Calvin Yee
Kernaussagen lesenHerausgeber öffnen (wird in einem neuen Tab geöffnet)

Auf einen Blick

  • 7% token cost reduction without quality loss

    Improvements to Cursor's agent harness across multiple layers reduced user token costs by 7% while maintaining agent quality.

    Unterstützendes Moment lesen · Absatz 7

  • 66% system prompt reduction enabled by model capability gains

    As models improved, explicit prohibitions and prescriptive instructions became unnecessary; Cursor trimmed approximately 66% of its system prompt by relying on model compliance with concise tool behavior definitions.

    Unterstützendes Moment lesen · Absatz 11

  • 46.9% token reduction from dynamic MCP tool loading

    Moving MCP tools into dynamic context—loading them only when invoked—reduced total tokens by 46.9% across sessions that used an MCP tool.

    Unterstützendes Moment lesen · Absatz 15

  • Cache-related changes reduced cold cache misses by 20%

    Cache-related changes reduced cold cache misses by 20%. The authors reserved tools and system instructions for content that rarely changes, then moved more variable setup past the cache boundaries into a “phantom user message.”

    Unterstützendes Moment lesen · Absatz 35

Wichtige Passagen4

Zugeordnete Passagen mit dem Kontext zur Überprüfung. Öffnen Sie den Originaltext, um die Quelle zu prüfen.

Absatz 7 (wird in einem neuen Tab geöffnet)
token efficiency optimization

7% token cost reduction without quality loss

Originalauszug

Changes across each of these layers reduced token costs for users by 7% without reducing agent quality.
Jediah Katz, Connor O’Keefe & Calvin Yee·
Kontext

Over the past few months we've responded to this shift by improving the efficiency of Cursor's agent harness. The harness gives us direct control over how each request is assembled, how context is reused, and when work is divided across agents.

Absatz 11 (wird in einem neuen Tab geöffnet)
system prompt design

66% system prompt reduction enabled by model capability gains

Originalauszug

This was true across model families, allowing us to trim roughly 66% of our system prompt.
Jediah Katz, Connor O’Keefe & Calvin Yee·
Kontext

As models improved, much of that direction became unnecessary. Instead of long lists of "DO NOT do this," "You must," or "Important" instructions, we could simply define how a tool behaves and models would generally comply.

Absatz 15 (wird in einem neuen Tab geöffnet)
dynamic context loading

46.9% token reduction from dynamic MCP tool loading

Originalauszug

This reduced total tokens by 46.9% across sessions that called an MCP tool.
Jediah Katz, Connor O’Keefe & Calvin Yee·
Kontext

That created an opportunity to improve efficiency by keeping tools available without including their full definitions in every request. We'd solved a similar problem earlier this year when we moved MCP tools into dynamic context, loading them only when needed.

Quelle & Methodik

Diese Standpunkte sind mit ihren Originalquellen verknüpft. Paraphrasen sind gekennzeichnet und keine wörtlichen Zitate.

Transkript oder Quellenmaterial öffnen (wird in einem neuen Tab geöffnet)Ein Problem melden