Maintainer of this skill?
Automated security scan — not a certification.
by phuryn
Bug Hunt Bench: 105 real bugs in two production repos, frontier coding models (GPT-6, Claude, Grok, Gemini, DeepSeek...) find and fix them in their own CLI, graded blind. Live leaderboard + every receipt.
# Add to your Claude Code skills
git clone https://github.com/phuryn/bug-hunt-benchGuides for using ai agents skills like bug-hunt-bench.
Last scanned: 9/14/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-09-14T09:35:33.605Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}See how bug-hunt-bench compares with popular alternatives.
bug-hunt-bench is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by phuryn. Bug Hunt Bench: 105 real bugs in two production repos, frontier coding models (GPT-6, Claude, Grok, Gemini, DeepSeek...) find and fix them in their own CLI, graded blind. Live leaderboard + every receipt. It has 176 GitHub stars.
Yes. bug-hunt-bench passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/phuryn/bug-hunt-bench" and add it to your Claude Code skills directory (see the Installation section above).
bug-hunt-bench is primarily written in JavaScript. It is open-source under phuryn on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh bug-hunt-bench against similar tools.
No comments yet. Be the first to share your thoughts!
From $49 · 30 days · terms & refunds
⚠️ Third-Party Software Notice
This skill is third-party open-source software developed and hosted independently on GitHub. SkillsLLM is an informational directory and does not control or maintain the underlying repository.
Any security checks, ratings, or warnings displayed by SkillsLLM are automated and limited in scope. They do not constitute a security certification or guarantee that the software is safe, error-free, or free from malicious code, vulnerabilities, compromised dependencies, or prompt-injection risks.
Review the source code, permissions, dependencies, and configuration before installing or running any third-party skill. Use is at your own risk. To the maximum extent permitted by applicable law, SkillsLLM is not liable for losses arising from third-party software.
Automated security scan — not a certification.
105 real bugs, hidden in two production codebases. Frontier coding models — GPT-6, Claude, Grok, Gemini, DeepSeek, Kimi, GLM and more — get one round per repo in their own agentic CLI (Codex CLI, Claude Code, Grok CLI, Antigravity CLI) to find and fix what they can. Every diff is graded blind against a withheld answer key. The score counts planted bugs only.
Live board: bughunt.productcompass.pm · Method, caveats and definitions · Raw data · Findings by wave
If the numbers save you a benchmark run of your own, star this repo — that is what keeps the bench findable, and new models are added as they ship.
Updated Oct 27, 2026 · 207 rows from 308 scored runs · 49 models · 25 of 105 bugs have never been fixed by any model.
Current leader: Sonnet 5.5 at max effort — 51.3 / 105 (22.7/45 on repo 1, 28.7/60 on repo 2).
Best per lab: Anthropic: Sonnet 5.5 (max) 51.3 · OpenAI: GPT-6 Astra (max) 45 · DeepSeek: DeepSeek V4.1 Flash 1017 (max) 42.7 · Meta: Muse Wonder Full 1.0 (max) 42 · Google: Gemini 4 Argon (high) 40.7 · Moonshot AI: Kimi K3.1 Pro (max) 36 · Unbiased: Pareto (ex-Union Alpha) (default) 30.7 · xAI: Grok 4.7 (xhigh) 29 · Alibaba: Qwen3.8-Flash (max) 26 · Xiaomi: MiMo-V2.6-Flash (default) 23.3 · Z.ai: GLM-5.3 (max) 19 · Tencent: Hy4 Preview (default) 18 · Mistral: Mistral Large 4 (high) 15
| # | Model | Harness | Effort | Runs | Fixed /105 | Repo 1 /45 | Repo 2 /60 | Extras | Wall | Cost | Date |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Sonnet 5.5 | Claude Code | max | 3 | 51.3 | 22.7 | 28.7 | 26.3 | 287 min | $154.03 | 2026-09-29 |
| 2 | GPT-6 Astra | Codex CLI | max | 3 | 45 | 22 | 23 | 55 | 90 min | $33.03 | 2026-09-14 |
| 3 | GPT-6.1 Sol | Codex CLI | max | 3 | 44.3 | 22.3 | 22 | 65.3 | 139 min | $5.88 | 2026-09-30 |
| 4 | GPT-5.6 Sol | Codex CLI | max | 2 | 43.5 | 19 | 24.5 | 48.5 | 253 min | $95.35 | 2026-09-16 |
| 5 | GPT-6 Astra | Codex CLI | xhigh | 1 | 43 | 23 | 20 | 53 | 59 min | $24.22 | 2026-09-04 |
| 6 | Fable 5.1 | Claude Code | max | 1 | 43 | 19 | 24 | 11 | 73 min | $87.18 | 2026-09-01 |
| 7 | DeepSeek V4.1 Flash 1017 | DeepSeek Harness / DeepSeek API | max | 7 | 42.7 | 19 | 23.7 | 44 | 62 min | $1.84 bill | 2026-10-18 |
| 8 | GPT-6.1 Sol | Codex CLI | xhigh | 3 | 42.7 | 22.3 | 20.3 | 56.7 | 99 min | $4.35 | 2026-09-30 |
| 9 | DeepSeek V4.1 Flash 1017 | Pi / DeepSeek API | max | 5 | 42.2 | 17.8 | 24.4 | 41.8 | 54 min | $0.83 bill | 2026-10-18 |
| 10 | Muse Wonder Full 1.0 | Muse Code / Meta API | max | 2 | 42 | 21.5 | 20.5 | 36.5 | 145 min | — | 2026-10-14 |
| 11 | Opus 5.5 | Claude Code | max | 3 | 41.7 | 19 | 22.7 | 9 | 67 min | $58.53 | 2026-09-23 |
| 12 | Muse Wonder Full 1.0 | Muse Code / Meta API | xhigh | 2 | 41.5 | 16 | 25.5 | 24.5 | 90 min | — | 2026-10-14 |
| 13 | GPT-6.1 Terra | Codex CLI | max | 3 | 40.7 | 18 | 22.7 | 48 | 37 min | $3.88 | 2026-10-26 |
| 14 | Gemini 4 Argon | Antigravity CLI | high | 3 | 40.7 | 13 | 27.7 | 7.3 | 56 min | $7.16 | 2026-10-14 |
| 15 | DeepSeek V4.1 Flash 1017 | Pi / DeepSeek API | high | 5 | 40.4 | 16 | 24.4 | 29.4 | 48 min | $0.64 bill | 2026-10-18 |
| 16 | Gemini 4 Argon | Antigravity CLI | medium | 3 | 39.3 | 13.7 | 25.7 | 11.3 | 60 min | $7.50 | 2026-10-14 |
| 17 | DeepSeek V4.1 Flash 1017 | DeepSeek Harness / DeepSeek API | high | 5 | 39.2 | 17 | 22.2 | 43.8 | 53 min | $1.27 bill | 2026-10-18 |
| 18 | DeepSeek V4.1 Flash 1017 | Claude Code / DeepSeek API | max | 7 | 39.1 | 18.1 | 21 | 22.9 | 58 min | $1.01 bill | 2026-10-18 |
| 19 | GPT-5.6 Sol | Codex CLI | xhigh | 1 | 39 | 18 | 21 | 59 | 126 min | $52.75 | 2026-08-28 |
| 20 | DeepSeek V4.1 Flash 1017 | OpenCode / DeepSeek API | max | 5 | 37.6 | 15.8 | 21.8 | 36.2 | 56 min | $0.96 bill | 2026-10-18 |
| 21 | DeepSeek V4.1 Flash 1017 | OpenCode / DeepSeek API | high | 5 | 36.6 | 15.4 | 21.2 | 25.6 | 42 min | $0.79 bill | 2026-10-19 |
| 22 | GPT-6.1 Sol | Codex CLI | high | 2 | 36.5 | 17 | 19.5 | 40 | 70 min | $3.26 | 2026-09-30 |
| 23 | DeepSeek V4.1 Flash 1017 | Claude Code / DeepSeek API | high | 5 | 36.4 | 15.2 | 21.2 | 24.4 | 54 min | $0.82 bill | 2026-10-17 |
| 24 | Kimi K3.1 Pro | Claude Code / Moonshot API | max | 3 | 36 | 17 | 19 | 29.3 | 65 min | $13.76 bill | 2026-10-23 |
| 25 | Opus 5.5 | Claude Code | xhigh | 3 | 36 | 17 | 19 | 13.3 | 42 min | $34.98 | 2026-09-23 |
| 26 | Sonnet 5.5 | Claude Code | xhigh | 3 | 36 | 16 | 20 | 14 | 73 min | $42.86 | 2026-09-29 |
| 27 | Gemini 4 Argon | Antigravity CLI | low | 3 | 35.3 | 12.7 | 22.7 | 5.7 | 60 min | $6.47 | 2026-10-14 |
| 28 | GPT-6 Astra | Codex CLI | high | 1 | 35 | 19 | 16 | 40 | 40 min | $20.60 | 2026-09-04 |
| 29 | DeepSeek V4.1 Flash 1017 | DeepSeek Harness / DeepSeek API | low | 5 | 34.8 | 16.4 | 18.4 | 34.4 | 34 min | $1.06 bill | 2026-10-18 |
| 30 | DeepSeek V4.1 Flash 1017 | Claude Code / DeepSeek API | low | 5 | 34.4 | 16.4 | 18 | 16.8 | 44 min | $0.71 bill | 2026-10-17 |
| 31 | Sonnet 5.5 | Claude Code | high | 3 | 34.3 | 15.7 | 18.7 | 10.7 | 28 min | $19.85 | 2026-09-29 |
| 32 | GPT-6 Astra | Codex CLI | medium | 1 | 34 | 19 | 15 | 33 | 28 min | $15.78 | 2026-09-05 |
| 33 | GPT-5.6 Sol | Codex CLI | high | 1 | 34 | 13 | 21 | 28 | 67 min | $33.92 | 2026-07-31 |
| 34 | DeepSeek V4.1 Flash 1017 | Pi / DeepSeek API | low | 5 | 33.8 | 14 | 19.8 | 19.6 | 33 min | $0.46 bill | 2026-10-18 |
| 35 | Fable 5.1 | Claude Code | high | 1 | 33 | 15 | 18 | 7 | 36 min | $48.58 | 2026-09-01 |
| 36 | DeepSeek V4.1 Flash 1017 | OpenCode / DeepSeek API | low | 5 | 32.8 | 15.8 | 17 | 17.6 | 41 min | $0.66 bill | 2026-10-19 |
| 37 | Muse Spark 1.3 | Muse Code / Meta API | max | 5 | 32.2 | 14.2 | 18 | 24 | 86 min | $18.11 | 2026-09-17 |
| 38 | GPT-5.6 Terra | Codex CLI | max | 1 | 32 | 16 | 16 | 45 | 160 min | $27.98 | 2026-08-27 |
| 39 | Opus 5.5 | Claude Code | high | 3 | 31.7 | 13 | 18.7 | 10.7 | 24 min | $22.25 | 2026-09-23 |
| 40 | Muse Wonder 1.0 | Muse Code / Meta API | max | 3 | 31.7 | 12 | 19.7 | 13.3 | 96 min | $27.87 | 2026-10-13 |
| 41 | GPT-5.6 Luna | Codex CLI | max | 3 | 31.3 | 16 | 15.3 | 54.7 | 187 min | $3.15 | 2026-09-22 |
| 42 | Pareto (ex-Union Alpha) | Claude Code / OpenRouter | default | 3 | 30.7 | 16 | 14.7 | 10.3 | 38 min | $4.81 | 2026-09-17 |
| 43 | Opus 5.5 | Claude Code | medium | 3 | 30.3 | 12.7 | 17.7 | 13.3 | 17 min | $15.68 | 2026-09-23 |
| 44 | GPT-6 Sol | Codex CLI | max | 3 | 29.3 | 15.3 | 14 | 41.3 | 63 min | $9.33 | 2026-09-23 |
| 45 | GPT-6.1 Sol | Codex CLI | medium | 2 | 29 | 16 | 13 | 30.5 | 30 min | $1.76 | 2026-09-30 |
| 46 | GPT-5.6 Sol | Codex CLI | medium | 1 | 29 | 13 | 16 | 24 | 48 min | $15.77 | 2026-09-06 |
| 47 | Grok 4.7 | Grok Build CLI (ACP) | xhigh | 5 | 29 | 13 | 16 | 32.2 | 45 min | $22.96 floor | 2026-10-08 |
| 48 | Fable 5.1 | Claude Code | low | 1 | 29 | 13 | 16 | 6 | 33 min | $33.00 | 2026-09-02 |
| 49 | Fable 5.1 | Claude Code | xhigh | 1 | 29 | 13 | 16 | 4 | 60 min | $53.61 | 2026-09-10 |
| 50 | Fable 5 | Claude Code | max | 1 | 29 | 12 | 17 | 5 | 57 min | $112.40 | 2026-08-01 |
| 51 | Grok 4.6 | Grok Build CLI (ACP) | xhigh | 3 | 28.7 | 11.7 | 17 | 18 | 43 min | $18.60 floor | 2026-09-14 |
| 52 | GPT-6 Astra | Codex CLI | low | 1 | 27 | 18 | 9 | 25 | 32 min | $11.69 | 2026-09-05 |
| 53 | Opus 5 | Claude Code | max | 1 | 27 | 13 | 14 | 2 | 60 min | $54.94 | 2026-08-01 |
| 54 | Grok 4.7 | Grok Build CLI (ACP) | medium | 3 | 26.7 | 12 | 14.7 | 22.7 | 31 min | $16.76 floor | 2026-09-21 |
| 55 | Qwen3.8-Flash | Claude Code / Alibaba API | max | 1 | 26 | 13 | 13 | 7 | 97 min | $1.81 | 2026-09-11 |
| 56 | Opus 5 | Claude Code | xhigh | 1 | 26 | 14 | 12 | 3 | 50 min | $63.25 | 2026-08-28 |
| 57 | Qwen3.8-Max | Claude Code / Alibaba API | max | 3 | 25.7 | 12.3 | 13.3 | 7 | 117 min | $26.76 | 2026-09-15 |
| 58 | GPT-6 Sol | Codex CLI | xhigh | 1 | 25 | 10 | 15 | 24 | 51 min | $7.67 | 2026-09-23 |
| 59 | Opus 5 | Claude Code | medium | 1 | 24 | 11 | 13 | 4 | 30 min | $37.40 | 2026-08-27 |
| 60 | Fable 5 | Claude Code | high | 1 | 24 | 9 | 15 | 3 | 31 min | $74.52 | 2026-07-26 |
| 61 | MiMo-V2.6-Flash | Claude Code / OpenRouter | default | 3 | 23.3 | 8 | 15.3 | 7.3 | 71 min | $0.49 bill | 2026-09-22 |
| 62 | Sonnet 5.5 | Claude Code | medium | 3 | 23.3 | 8.7 | 14.7 | 8.7 | 17 min | $11.11 | 2026-09-30 |
| 63 | Qwen3.8-Flash | Claude Code / Alibaba API | low | 1 | 23 | 11 | 12 | 7 | 98 min | $1.37 | 2026-09-11 |
| 64 | GPT-5.6 Luna | Codex CLI | xhigh | 1 | 23 | 10 | 13 | 53 | 136 min | $2.50 | 2026-08-28 |
| 65 | Grok 4.6 | Grok Build CLI (ACP) | high | 1 | 23 | 7 | 16 | 16 | 33 min | $15.73 floor | 2026-08-28 |
| 66 | MiMo-V2.6-Pro | Claude Code / OpenRouter | default | 3 | 22.7 | 11.3 | 11.3 | 5.3 | 91 min | $0.86 bill | 2026-09-22 |
| 67 | GPT-6.1 Sol | Codex CLI | low | 2 | 22.5 | 11.5 | 11 | 24 | 25 min | $1.34 | 2026-10-01 |
| 68 | Opus 5.5 | Claude Code | low | 3 | 22.3 | 11.3 | 11 | 9.7 | 12 min | $8.34 | 2026-09-23 |
| 69 | Grok 4.6 | Grok Build CLI (ACP) | medium | 1 | 22 | 9 | 13 | 11 | 26 min | $5.80 floor | 2026-08-28 |
| 70 | Gemini 3.7 Flash | Gemini CLI (retired) + model-pinning gateway | high | 1 | 22 | 8 | 14 | 4 | 97 min | $8.43 | 2026-08-24 |
| 71 | DeepSeek V4.1 Flash | Claude Code / DeepSeek API | max | 3 | 21.7 | 11.3 | 10.3 | 6.7 | 56 min | $0.78 bill | 2026-09-15 |
| 72 | Haiku 5.5 | Claude Code | max | 2 | 21.5 | 10 | 11.5 | 9 | 32 min | $5.83 | 2026-10-07 |
| 73 | Fable 5.1 | Claude Code | m |