cybergym-e2e

CyberGym-E2E is a large-scale benchmark built from real-world vulnerabilities in widely used open-source projects to evaluate AI agents' end-to-end cybersecurity capabilities, from discovering vulnerabilities to generating proof-of-concept to writing patches.

sunblaze-ucb/cybergym-e2e

View on GitHub ↗

CyberGym-E2E is a large-scale benchmark built from real-world vulnerabilities in widely used open-source projects to evaluate AI agents' end-to-end cybersecurity capabilities, from discovering vulnerabilities to generating proof-of-concept to writing patches.

cybergym-e2e grew +25.0% in the last 30 days, ranks 1151st of 1,388 in Domain-Specific Apps, actively maintained, last commit 3 days ago.

Star trajectory

80 days tracked
2026-09-02110 stars today2026-11-22
Momentum

+25.0% in 30d, above the Domain-Specific Apps median of +3.3%.

Maintenance

Last commit 3 days ago, fresher than the median of 13d.

How it compares in Domain-Specific Apps

cybergym-e2e
110
stars
+25.0% 30d
commit 3 days ago
Domain-Specific Apps median
257.5
stars
+3.3% 30d
commit ~13d ago
Leader · serviette
58.7k
stars
-0.2% 30d
commit 11 days ago

Ranked 1151st of 1,388 by stars.

Ranked by shared tags, then keywords, then stars
  1. ExploitGym is a large-scale, realistic benchmark built from real-world vulnerabilities designed to evaluate AI…

    +114 (+9.5%) in 30d · Committed 8d ago · C · Apache-2.0

    Shares Domain-Specific Apps

  2. A modular framework for benchmarking LLMs and agentic strategies on security challenges across HackTheBox…

    +5 (+1.1%) in 30d · Committed 120d ago · Python · AGPL-3.0

    Shares Domain-Specific Apps

  3. Visa Vulnerability Agentic Harness

    +97 (+3.2%) in 30d · Committed yesterday · Python · Other license

    Shares Domain-Specific Apps

  4. Terminal-Bench-Science: Evaluating AI agents on research workflows across scientific domains

    +203 (+27%) in 30d · Committed 3d ago · Python · Apache-2.0

    Shares Domain-Specific Apps

  5. 检查第三方AI API是否掺假以及渠道一致|基于llm指纹的AI模型识别

    +59 (+7.1%) in 30d · Committed 2d ago · HTML · LGPL-2.1

    Shares Domain-Specific Apps

  6. The open source visual testing platform for teams and AI agents. Review the product, not just the code.

    +20 (+3.1%) in 30d · Committed today · TypeScript · MIT

    Shares Domain-Specific Apps

  7. Agentic QE Fleet is an open-source AI-powered QA/QE platform designed for use with Coding Agents (works best…

    +19 (+3.7%) in 30d · Committed today · TypeScript · MIT

    Shares Domain-Specific Apps

  8. Design, conduct and analyze results of AI-powered surveys and experiments. Simulate social science and market…

    +14 (+2.8%) in 30d · Committed today · Python · MIT

    Shares Domain-Specific Apps

  9. An autonomous AI scientist: a multi-agent loop over literature, experiments, self-critique and write-up, with…

    +6 (+1.3%) in 30d · Committed 5d ago · Python · MIT

    Shares Domain-Specific Apps

  10. MARVIS-Agent: all-purpose credit risk agent for model development, validation, data processing, feature…

    +7 (+1.8%) in 30d · Committed 12d ago · Python · MIT

    Shares Domain-Specific Apps

Language
Shell
License
Apache License 2.0
Contributors
1
Forks
30
Watchers
1
Open issues
0
Releases
0
Created
2026-05-14
Stars this week
+4

You've seen where cybergym-e2e stands today.

Get the moves before everyone else: the 5 Domain-Specific Apps repos climbing fastest, every Tuesday.

Free weekly digest. Double opt-in, unsubscribe anytime.

Maintain cybergym-e2e?

Claim this page to add your own call-to-action.

Claim this page