Skip to content
View Gideon145's full-sized avatar
:octocat:
:octocat:

Block or report Gideon145

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Gideon145/README.md

Hi, I'm Gideon πŸ‘‹

Cybersecurity Engineer β€” MSc Cyber Security, University of Aberdeen. I build detection and response tooling at the intersection of AI security, threat intelligence, and protocol design.

Email LinkedIn


What I Build

  • AI & LLM Security β€” Prompt injection firewalls, genetic algorithm fuzzers, autonomous SOC analysts
  • Detection Engineering β€” Sigma rules in Splunk, Sentinel, QRadar, CrowdStrike; YARA & Snort signatures
  • Blockchain Protocol Security β€” Audited smart contracts, multi-agent oracles, on-chain economic security

Featured Projects

Project Description
Prompt Firewall HTTP proxy blocking LLM prompt injection. 30+ signatures, 100% detection, 0 false positives
AI SOC Analyst Autonomous Tier-1 analyst β€” log ingestion, MITRE ATT&CK mapping, auto-correlation
InjectionForge Genetic algorithm fuzzer β€” discovered 7 novel LLM jailbreak bypass techniques
Argus Multi-agent security oracle β€” 350+ users, ChainGPT 5/5 audit, top 5 ASP on OKX.AI
Mirai Detection Toolkit YARA rules, Snort signatures, C2 simulator, IoT credential scanner
SigmaHQ Rules Mirai C2 detection rules reviewed by SigmaHQ maintainers Β· Elastic rule under review (PR #6647)
TrustGuard ELO reputation + risk scoring β€” 11 adversarial tests, peer-reviewed
Breach Postmortems Independent IR analyses: Levi Strauss, Swiss Gov, UNC6671, ChainDrop, Claude AI

Recognition

  • πŸ₯‰ OKX Build X Season 2 β€” 3rd Place, X Layer Arena
  • πŸ”¬ MIT Media Lab Contributor
  • πŸ›‘οΈ SigmaHQ-Reviewed & Elastic Detection Rule Contributor
  • βœ… ChainGPT 5/5 Security Audit
  • πŸ“‘ OKX.AI Pioneer β€” 6 autonomous security agents

Currently

  • MSc Cyber Security at University of Aberdeen (est. 1495)
  • Seeking security engineering roles β€” detection, AI security, threat intelligence
  • Based in Aberdeen, Scotland, UK

Building tools that actually work.

Pinned Loading

  1. argus argus Public

    Arc's first multi-agent security oracle. 3 AI agents stake USDC on verdicts. 360+ users, 1,637+ scans. 5/5 Circle primitives. ChainGPT audited. An official Circle Alliance member.

    TypeScript 5 1

  2. detection-engineering-lab detection-engineering-lab Public

    Full-cycle detection engineering lab: Sigma rules, ATT&CK-mapped investigations, incident response workflow, and rule tuning. Built on the Elastic Stack with real Windows event logs.

  3. elk-siem-lab elk-siem-lab Public

    Full ELK SIEM stack in Docker: 500 synthetic security events, custom detection rules, and Kibana dashboards for hands-on SOC practice.

    Python

  4. injection-forge injection-forge Public

    Genetic-algorithm fuzzer that evolves prompt-injection attacks across generations to discover LLM guardrail bypasses. Found 7 novel evasion techniques.

    Python

  5. llm-prompt-firewall llm-prompt-firewall Public

    HTTP proxy that blocks prompt injection, jailbreak, and data-exfiltration attacks before they reach an LLM. 30+ signatures, 100% detection across 62 adversarial cases.

    Python