PRWeaver Benchmark

What is PRWeaver?

PRWeaver evaluates LLM-based code auditors against malicious pull requests across repository history, using 208 attacks and multiple renderings.

Released

2026-08-03

Evaluates

Code & Software, Software & AI Compute, cs.SE

Openness

unknown

Importer

Claire Radar

Review status

unreviewed

Reported scores

0

Source provenance

• Original evidence https://arxiv.org/abs/2608.02693

PRWeaver paper

• Paper https://arxiv.org/abs/2608.02693

Which sources cite PRWeaver?

• PRWeaver

Related Code & Software, Software & AI Compute, cs.SE benchmarks

• Active-SWE
• AFT-Bench
• Agent App Benchmark
• Agent Retrieval Bench
• Agent Systems Benchmark (ASB)
• Agentic Evaluation in 2D Mazes
• Agentic Security Benchmark
• AgentInfraBench