Skip to content

Wire Briefings Search

Fetching from the wire…

01

02

03

04

05

06

07

08

Wire Briefings Search

Entity trail

Recuris

Source-backed findings, relationship evidence, citations, and briefing history from the public MindPattern archive.

Briefing refs

1

Findings

1

Edges

4

Sources

1

Corpus findings

2026-08-26 / skill-finderSplitting agent memory into working and experiential halves lifts tau-bench to 87.9% on Claude Opus 5, a 15.6-point gainRecuris (arXiv 2608.24876, submitted 2026-08-25) keeps a Working Memory that tracks current task progress separate from an Experiential Memory of learned skills, so skill selection is grounded in what the task needs right now instead of the full history. It improves success in 35 of 37 model-benchmark pairs, gains +17.8 points on GPT-5.6 Sol, reaches +32.2 points on the longest-horizon tasks, and cuts common failure modes by up to 80% via a Meta-Agent that localizes failures to a specific memory component and applies validation-gated skill updates. For anyone running long-horizon harnesses, the actionable part is the split itself: a progress tracker that indexes into a skill store beats appending everything to one growing context.

Graph relationships

BENCHMARKED AGAINST
MBPP
Recuris improves on MBPP+ benchmarks.
Source finding
BENCHMARKED AGAINST
HumanEval
Recuris improves on HumanEval+ benchmarks.
Source finding
USES
Claude Opus 5
Recuris improves success on Claude Opus 5 by 15.6 points.
Source finding
BENCHMARKED AGAINST
Tau-bench
Recuris lifts tau-bench to 87.9% on Claude Opus 5.
Source finding

Source trail

arXiv 2608.24876

Graph sources

entity graphfindings textkg edgeskg entitiesnewsletter issues