# Task Synthesis

GraphForge is an evidence-graph framework that starts from O*NET occupation seeds, crawls real files into a workspace, builds an evidence graph over their relations, and compiles that graph into task statements and rubrics, with an initial rollout and a revision agent checking executability; fine-tuning Qwen3.6-27B on 2,169 trajectories brings GDPVal to 1445.7 (+65.7) under OpenHands and Workspace-Bench-Lite and SpreadsheetBench II to 63.7 (+7.7) and 24.0 (+13.7) under Claude Code, with the same data also improving Qwen3.6-35B-A3B.

GraphForge is an evidence-graph framework that starts from O*NET occupation seeds, crawls real files into a workspace, builds an evidence graph over their relations, and compiles that graph into task statements and rubrics, with an initial rollout and a revision agent checking executability; fine-tuning Qwen3.6-27B on 2,169 trajectories brings GDPVal to 1445.7 (+65.7) under OpenHands and Workspace-Bench-Lite and SpreadsheetBench II to 63.7 (+7.7) and 24.0 (+13.7) under Claude Code, with the same data also improving Qwen3.6-35B-A3B.

SkillGym proposes an automatic pipeline that crawls skills from the internet and keeps those whose workflows run reproducibly offline, uses a builder-reviewer agent system to construct skill-critical tasks across four reasoning structures each with a reference solution and an executable verifier, then collects successful trajectories from three teacher models and four agent harnesses for supervised finetuning, improving six models from 2B to 122B parameters on 22 of 24 comparisons across four skill-use benchmarks, with the 9B model outperforming Qwen3.5-397B-A17B on the SkillGym test set and SkillEval and raising the rate of reading the relevant skill from 28% to 96%.

SkillGym proposes an automatic pipeline that crawls skills from the internet and keeps those whose workflows run reproducibly offline, uses a builder-reviewer agent system to construct skill-critical tasks across four reasoning structures each with a reference solution and an executable verifier, then collects successful trajectories from three teacher models and four agent harnesses for supervised finetuning, improving six models from 2B to 122B parameters on 22 of 24 comparisons across four skill-use benchmarks, with the 9B model outperforming Qwen3.5-397B-A17B on the SkillGym test set and SkillEval and raising the rate of reading the relevant skill from 28% to 96%.

Skill2Env is a capability-oriented framework that curates skills from public skill libraries, uses reusable difficulty patterns and task blueprints to jointly synthesize task instructions, execution substrates, workspaces, and rubric-based evaluators, and iteratively hardens tasks using solver execution evidence, yielding 2,963 executable tasks; supervised fine-tuning of Qwen3.6-35B-A3B on 1.5K high-scoring trajectories from these environments raises the unweighted average across seven agent benchmarks from 36.6 to 45.0.

Skill2Env is a capability-oriented framework that curates skills from public skill libraries, uses reusable difficulty patterns and task blueprints to jointly synthesize task instructions, execution substrates, workspaces, and rubric-based evaluators, and iteratively hardens tasks using solver execution evidence, yielding 2,963 executable tasks; supervised fine-tuning of Qwen3.6-35B-A3B on 1.5K high-scoring trajectories from these environments raises the unweighted average across seven agent benchmarks from 36.6 to 45.0.