# Envs-FORGE: Frontier-Optimized Reward-Grounded Environment Synthesis for Agent RL

arXiv: https://arxiv.org/abs/2608.14312 · HTML: https://mlsanity.com/paper/2608.14312/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 10 of its 13 references that are in the map.
- First posted on arXiv 14 Aug 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 10 of 10)

- [Self-Instruct: Aligning Language Models with Self-Generated Instructions](https://mlsanity.com/paper/2212.10560/) (2022)
- [WizardLM: Empowering large pre-trained language models to follow complex instructions](https://mlsanity.com/paper/2304.12244/) (2023)
- [Terminal-Bench: Benchmarking Agents on Hard, Realistic Tasks in Command Line Interfaces](https://mlsanity.com/paper/2601.11868/) (2026)
- [AgentInstruct: Toward Generative Teaching with Agentic Flows](https://mlsanity.com/paper/2407.03502/) (2024)
- [Endless Terminals: Scaling RL Environments for Terminal Agents](https://mlsanity.com/paper/2601.16443/) (2026)
- [Large-Scale Terminal Agentic Trajectory Generation from Dockerized Environments](https://mlsanity.com/paper/2602.01244/) (2026)
- [TermiGen: High-Fidelity Environment and Robust Trajectory Synthesis for Terminal Agents](https://mlsanity.com/paper/2602.07274/) (2026)
- [CLI-Gym: Scalable CLI Task Generation via Agentic Environment Inversion](https://mlsanity.com/paper/2602.10999/) (2026)
- [Toward Scalable Terminal Task Synthesis via Skill Graphs](https://mlsanity.com/paper/2604.25727/) (2026)
- [Targeted synthetic data generation for tabular data via hardness characterization](https://mlsanity.com/paper/2410.00759/) (2024)

## Built on by (most cited first, 4 of 4)

- [FrogNano: Training a 4B Coding Agent via Online Task Synthesis](https://mlsanity.com/paper/2609.07925/) (2026)
- [V-Gym: Enhancing Agentic Visual Reasoning via Skill-Data Co-Evolution](https://mlsanity.com/paper/2609.34682/) (2026)
- [AutoEnvScaling: Automating the Data Flywheel with Terminal Agents](https://mlsanity.com/paper/2610.14163/) (2026)
- [WorkRover: Scaling Long-Horizon Cowork Agent Training Via Recursive Environment Evolution](https://mlsanity.com/paper/2610.18067/) (2026)

## Abstract (arXiv)

Reinforcement learning (RL) for terminal agents needs executable training environments with reliable rewards and useful difficulty. Fixed recipes such as few-shot, Self-Instruct, and Evol-Instruct apply the same prompting policy to every seed, even when the current policy would benefit from a harder, easier, or simply different task. We present Envs-FORGE, a prompting policy that converts verifier rewards into per-seed environment-synthesis actions. Envs-FORGE estimates seed pass rates, scores six projection--direction actions around a target learning frontier, and solves a per-seed mixed-integer linear program (MILP) to choose the action that conditions generation. The selected action drives synchronized rewriting of the instruction, fixtures, oracle solution, tests, and Docker environment; only gold-verified bundles enter RL training. The indexed MILP form also supports optional soft skill coverage for portfolio planning. On Qwen 3.5 35B, Envs-FORGE improves Pass@1 over Base by 9.2 percentage points on tb-core (40.0% to 49.2%) and 6.4 points on tb-2.0 (23.0% to 29.4%), exceeding the strongest fixed-recipe baseline by 2.4 and 2.1 points. It reaches 77.1% on SWE-bench Verified versus 73.4% for Base, and improves tb-core by 6.8--9.2 points across the evaluated 4B--35B models. All synthesis methods export 100 verified environments and use 2.27M--2.88M synthesis tokens, placing the comparison at the same downstream training-set size and the same operational scale. The source code is available at https://github.com/DataArcTech/DataArc-SynData-Toolkit/.