terminal-bench repositories

Trending repositories tagged terminal-bench, ranked by trending score. Updated daily.

  1. 1
    simple-agent-lab/RSIHub

    A research framework for principled agent self-improvement under frozen evaluators and declared mutation boundaries, recording verifiable lineage to make it reproducible and auditable.

    GitHub · Python · 379 stars · 20 forks · +156 stars in 24 hours · Trending score 3.67

  2. 2
    harbor-framework/harbor

    Framework for evaluating and improving agents

    GitHub · Python · 7,264 stars · 2,250 forks · 0 stars in 24 hours · Trending score 1.70

  3. 3
    UiPath/coder_eval

    Playwright for coding agents. Test that your skills, MCP servers, and CLIs actually work when an agent uses them — sandboxed YAML suites, A/B experiments, CI gates.

    GitHub · Python · 159 stars · 10 forks · +6 stars in 24 hours · Trending score 1.61

  4. 4
    tribgames/mixdog

    Open-source desktop coding agent where a lead model plans and cheaper models handle search, edits and review.

    GitHub · Rust · 259 stars · 29 forks · +2 stars in 24 hours · Trending score 1.24

  5. 5
    mmprotest/villani-code

    Some tools help. Some tools assist. Villani Code intervenes.

    GitHub · Python · 124 stars · 19 forks · 0 stars in 24 hours · Trending score 0.75

  6. 6
    alanhuangyoo/crux

    Taking the pi coding agent to Claude Code-level performance: 0.539 → 0.773 pass@1 on Terminal-Bench 2.1 with the same self-hosted 27B model, by re-engineering the agent.

    GitHub · TypeScript · 187 stars · 36 forks · +1 stars in 24 hours · Trending score 0.45

  7. 7
    macanderson/stella

    A coding agent that builds it's own tools and mines it's own skills and always ships with receipts.

    GitHub · Rust · 19 stars · 3 forks · 0 stars in 24 hours · Trending score 0.22

  8. 8
    todoforai/edge

    Open-source AI agent that runs on your own computer and does your tasks. Any model (Claude, GPT, Gemini, Grok, Ollama), drives your real browser, shell, files, DOCX/XLSX.

    GitHub · TypeScript · 87 stars · 1 forks · 0 stars in 24 hours · Trending score 0.07

  9. 9
    aktasbatuhan/compound

    Which host should serve your open model? Run your workload across pinned, verified providers and get one table: success, cost, latency, cache hits.

    GitHub · TypeScript · 8 stars · 0 forks · 0 stars in 24 hours · Trending score 0.00

  10. 10
    bugyal/skeptic

    Runs BOTH controls (nop + oracle) across EVERY task in a coding-agent benchmark, classifies the outcome, captures evidence, and lints leaks statically. One binary, zero LLM calls.

    GitHub · Go · 5 stars · 1 forks · 0 stars in 24 hours · Trending score 0.00

  11. 11
    quqxui/RSI-Flow

    A recursive self-improvement (RSI) workflow that evolves both target agents (e.g., Pi and DeepSeek Harness) and the RSI system itself.

    GitHub · Python · 23 stars · 0 forks · 0 stars in 24 hours · Trending score 0.00

  12. 12
    zli12321/LHTB

    Long Horizon Terminal Benchmark with Dense Reward Grading

    GitHub · Python · 570 stars · 35 forks · −145 stars in 24 hours · Trending score -0.02