Terminal-Wrench, a dataset of 331 realistic hackable environments
I want to share a new dataset of 331 reward-hackable environments. These are real environments used in Terminal Bench and adjacent benchmarks. I first got interested in this because, as a reviewer of Terminal Bench, I noticed a lot of our tasks were hackable. I also noticed that many contributors to the benchmark do so because it provides credibility when selling environments to labs. Hence, TBench tasks are, in my opinion, held to a higher quality standard than those being used today for RL. No one is spending hours manually reviewing the $1B in tasks being purchased by major labs. As far…
What it does
In the maker’s words, at launch
I want to share a new dataset of 331 reward-hackable environments. These are real environments used in Terminal Bench and adjacent benchmarks. I first got interested in this because, as a reviewer of Terminal Bench, I noticed a lot of our tasks were hackable. I also noticed that many contributors to the benchmark do so because it provides credibility when selling environments to labs. Hence, TBench tasks are, in my opinion, held to a higher quality standard than those being used today for RL. No one is spending hours manually reviewing the $1B in tasks being purchased by major labs. As far as I understand, while everyone knows environments are hackable, nobody has released hundreds of "realistic" environments.
Does a similar job
all alternatives →- OAOSS Agent I built topped the TerminalBench on Gemini-3-flash-previewApr 2026 · github.com · ▲393
Scored 65.2% vs google's official 47.8%, and the existing top closed source model Junie CLI's 64.3%. Since there are a lot of reports of deliberate cheating on TerminalBench 2.0 lately (https://debugml.github.io/cheating-agents/), I would like to also clarify a few things 1. Absolutely no {agents/skills}.md files were inserted at any point. No cheating mechanisms whatsoever 2. The cli agent was run in leaderboard compliant way (no modification of resources or timeouts) 3. The full terminal bench run was done using the fully open source version of the agent, no…
- TBTerminal-Bench-RL: Training long-horizon terminal agents with RL2025 · github.com · ▲125
After training calculator agent via RL, I really wanted to go bigger! So I built RL infrastructure for training long-horizon terminal/coding agents that scales from 2x A100s to 32x H100s (~$1M worth of compute!) Without any training, my 32B agent hit #19 on Terminal-Bench leaderboard, beating Stanford's Terminus-Qwen3-235B-A22! With training... well, too expensive, but I bet the results would be good! *What I did*: - Created a Claude Code-inspired agent (system msg + tools) - Built Docker-isolated GRPO training where each rollout gets its own container - Developed a multi-agent…
- FAFresh – A new terminal editor built in RustDec 2025 · sinelaw.github.io · ▲187
I built Fresh to challenge the status quo that terminal editing must require a steep learning curve or endless configuration. My goal was to create a fast, resource-efficient TUI editor with the usability and features of a modern GUI editor (like a command palette, mouse support, and LSP integration). Core Philosophy: - Ease-of-Use: Fundamentally non-modal. Prioritizes standard keybindings and a minimal learning curve. - Efficiency: Uses a lazy-loading piece tree to avoid loading huge files into RAM - reads only what's needed for user interactions. Coded in Rust. - Extensibility: Uses…
hackerDenMay 2026 · hackerden.netlify.app · ▲77Collaborative hackathon workspace with visible contributions
- RURewardHackBench: Using sandboxes to stop agents from cheatingJun 2026 · github.com · ▲9
hey all, happy to share research i've been working on for islo.dev in recent months. ever since the cheating agents (https://debugml.github.io/cheating-agents/) paper came out, revealing reward hacking was 4x more prevalent than previously estimated, i've been looking into how we can deal with the issue the common approach (taken by the tbench team) is post hoc trajectory analysis. i've been interested in the idea of reframing the problem as an endpoint security problem and tackling it via sandbox i hope you find it interesting, and thanks to the islo.dev team for…
- MAMy Attempt to Organize the World of AI Dev Tools2025 · aicode.danvoronov.com · ▲101
I've been exploring the (not so=) amazing potential of AI in coding and have compiled a list of tools. From AI-powered IDEs to code generators, this resource is my contribution to the community. I'm still on the fence about including txt2sql projects, as their functionality seems too basic to me. And I'm personally maintaining this, so your feedback is wellcome.
More dev tools this month
the category →
OpenTrailPaper is open-source bike computer firmware for the LilyGO T5S3 4.7" E-Paper PRO. It supports offline maps, GPX routes, FIT recording and Bluetooth sensors.
Dev tools · 2d ago · opentrailpaper.com
Open-source GTM skills for technical founders
Dev tools · 30d ago · gtmcofounder.com
Launched alongside, April 2026
the whole month →
- AG
Thought the resources for GPU arch were lacking, so here we are
Life & fun · Apr 2026 · jaso1024.com
- IB
Built a ~9M param LLM from scratch to understand how they actually work. Vanilla transformer, 60K synthetic conversations, ~130 lines of PyTorch. Trains in 5 min on a free Colab T4. The fish thinks the meaning of life is food. Fork it and swap the personality for your own character.
AI · Apr 2026 · github.com
- BC
Life & fun · Apr 2026 · sam-burns.com
- IB
With social media and now AI, its important to keep the indie web alive. There are many people who write frequently. Blogosphere tries to highlight them by fetching the recent posts from personal blogs across many categories. There are two versions: Minimal (HN-inspired, fast, static): https://text.blogosphere.app/ Non-minimal: https://blogosphere.app/ If you don't find your blog (or your favorite ones), please add them. I will review and approve it.
AI · Apr 2026 · text.blogosphere.app