Search results

Question: Why does the judge truncate files so aggressively?

Status: Open.

#96 In pinchbench/skill;

· thomas-tharp opened on Apr 2, 2026

Task 1 incorrectly allows incorrect times to pass

Status: Open.

#94 In pinchbench/skill;

· xeophon opened on Apr 2, 2026

[Feature Request] 'Vibe' Benchmark

Status: Open.

#92 In pinchbench/skill;

· joshavant opened on Apr 2, 2026

Why task_11 can test the ClawHub interaction?

Status: Open.

#80 In pinchbench/skill;

· Chaotic-chaos opened on Mar 26, 2026

LLM Judge Prompt Design

Status: Open.

#75 In pinchbench/skill;

· PlanckChang opened on Mar 25, 2026

benchmark: wait for transcript completion instead of assuming subprocess exit means task finished

Status: Open.

#65 In pinchbench/skill;

· jiaxin576 opened on Mar 18, 2026

Multi-domain benchmarking: AgentBench-Live covers 5 domains beyond coding

Status: Open.

#64 In pinchbench/skill;

· jackjin1997 opened on Mar 18, 2026

Meta: Migrate from Git Hash Versioning to Semantic Versioning

enhancementNew feature or requestNew feature or request

Status: Open.

#62 In pinchbench/skill;

· ScuttleBot opened on Mar 17, 2026

Meta: PinchBench v2

Status: Open.

#60 In pinchbench/skill;

· olearycrew opened on Mar 17, 2026

Leaderboard scoring bias: runs with fewer tasks rank higher than comprehensive runs

Status: Open.

#53 In pinchbench/skill;

· JoeProAI opened on Mar 15, 2026

Discussion: What tasks should we add next?

Status: Open.

#52 In pinchbench/skill;

· ScuttleBot opened on Mar 15, 2026

Fix or remove task that fails 95% of the time on 90% of models

Status: Open.

#51 In pinchbench/skill;

· ScuttleBot opened on Mar 15, 2026