SE Bug Bounty Arena / Hunter

Maryam Mohamed

Exploit HunterLevel 1729 of 104 labs solved0 reports resolved

Hunter rating: Hunter

A score from 0 to 100 for current AI bug hunting ability. It rises with harder labs, better reports and wider coverage, and it falls after 30 days without activity.

  • 32 / 45Labs solved, weighted by level
  • 0 / 25Report quality
  • 13 / 15Bug classes covered
  • 0 / 15Exam, challenge and operations

Last active 17 Nov 2026.

✓ Every solve, score and certificate on this page was recorded by SecurityElites. Nothing here is self-reported.

Coverage of the OWASP Top 10 for LLM Applications (2025)

Labs solved in each category this Arena covers.

  • LLM01 Prompt Injection11/22
  • LLM02 Sensitive Information Disclosure1/3
  • LLM03 Supply Chain1/4
  • LLM04 Data and Model Poisoning2/4
  • LLM05 Improper Output Handling1/8
  • LLM06 Excessive Agency4/16
  • LLM07 System Prompt Leakage3/5
  • LLM08 Vector and Embedding Weaknesses3/4

Badges

  • First bugSolved a first Arena lab
  • Ten downSolved 10 Arena labs
  • Cleared AtlasmindFound every bug on the Atlasmind target

Bugs found by class

  • Prompt injection4/8
  • Jailbreaks4/8
  • System prompt leakage3/5
  • Data leaks1/3
  • Unsafe output handling1/8
  • AI-written code0/3
  • Indirect injection3/6
  • RAG and embeddings3/4
  • Agent and tool hijacking2/8
  • Excessive agency2/8
  • Agent memory2/4
  • MCP and supply chain1/4
  • Synthetic AI and model attacks0/5
  • Web bugs behind AI apps3/30
Share this profile:LinkedInX

Start hunting in the Arena