Unverified50% confidenceEventExact time
By removing half the tools and tightening the schema definitions, a developer noticeably improved the Agent tool-calling accuracy of GPT-5.4 and Claude Opus, while increasing prompt length did not improve the behavior.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/18/2026
First Seen
Valid until: 1/16/2027
Sources
Related Entities
Related Claims
UnverifiedThe developer's Agent was connected to Slack, Notion, HubSpot and internal APIs; GPT-5.4 frequently called the wrong functions, and after switching to Claude Opus the problem persisted but improved slightly.75% similarVerifiedThe instruction-following and context comprehension capabilities of new-generation models such as GPT-4, Claude 3.5, and DeepSeek-V3 have improved significantly, making complex prompt templates no longer necessary.74% similarUnverifiedIn the optimization workflow of web-edit 0.1a1, GPT-6 Astra was responsible for diagnosing performance bottlenecks, Claude Opus 5.5 was responsible for rewriting the code, and Simon Willison was responsible for reviewing, merging and releasing the results.70% similarUnverifiedIt is estimated that GPT-4's tool-calling capability alone underwent thousands of hours of manual annotation and multiple rounds of iterative optimization.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1059140API
curl https://kongchang.com/api/v1/knowledge/claims/1059140MCP
get_claim(id=1059140)