Skip to content
View gesh75's full-sized avatar

Highlights

  • Pro

Block or report gesh75

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.

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
gesh75/README.md

Georgi Gaydarov

Senior Network Engineering Leader (20+ yrs) · AI Tools Builder

Multivendor networks (Juniper · Arista · Cisco · Palo Alto · Fortinet) · BGP/MPLS · SD-WAN · SASE · Zero Trust · Python automation · on-prem & agentic LLMs.

The working name for the stack is GESH.


Docs hub → gesh75.github.io

Live one-page docs (with architecture diagrams) for my multivendor network automation, AI-for-NetOps, and security projects.

System What it is Docs
multivendor-ai-network-lab 26-device lab that closes the loop — Pydantic-AI orchestrator, 68 MCP tools, risk-gated auto-remediation, RFC 6241 confirmed-commit rollback, immutable GAIT audit. docs
argus Agentic AI pentester — reasons, chains & adapts across network/host/AD/web, read-only behind a fail-closed 7-layer guardrail. HMAC-audited. Claude / Ollama / offline. docs
aegis Air-gapped pre-deployment change validation against a real containerlab digital twin. Sealed PCI/SOC2/NIST evidence. Zero egress. docs
multivendor-cli-configurator 70,000+ CLI commands across 17 vendors — searchable N-vendor compare + NETCONF/Ansible drawer. Single HTML, zero deps. docs
netlog-ai Sanitize-first AI log analyzer (Junos / EOS / FRR). LLM root-cause playbooks. 139 tests. docs
napalm-live-lab Honest, live NAPALM coverage matrix + safe-by-default command console — Arista cEOS / Nokia SR Linux / FRR. docs
network-observability-architecture Vendor-neutral reference architecture for network observability and source-of-truth management. docs
claude-mastery-hub Interactive guide to Claude — app, Claude Code, API, MCP, skills, subagents & hooks. docs
claude-skill-lint Zero-dependency linter for Claude Code skills — frontmatter, progressive disclosure, dead refs, stale model IDs. source

Doctrine: fail closed · sanitize first · evidence leaves, packets don’t · honest coverage.


Tech

Juniper Arista Cisco Palo Alto Fortinet FRR VyOS · BGP/MPLS SD-WAN SASE Zero Trust
Python FastAPI Ansible Nornir Containerlab · Anthropic Claude Pydantic-AI on-prem LLMs · Docker

Connect

GitHub · LinkedIn · X · Docs

Popular repositories Loading

  1. multivendor-ai-network-lab multivendor-ai-network-lab Public

    26-device multivendor AI network lab — now closes the loop: detected anomalies auto-remediate, risk-gated (auto / approve / page-out) with RFC 6241 confirmed-commit rollback. Pydantic-AI orchestrat…

    Python 11 7

  2. multivendor-cli-configurator multivendor-cli-configurator Public

    70,000+ CLI commands across 17 network vendors & tools (Cisco, Juniper, Arista, FRR, VyOS, Huawei, Aruba, Fortinet, Palo Alto, Nokia, Mikrotik, Extreme, NVIDIA, SONiC + PowerShell/Linux/Wireshark).…

    HTML 11 2

  3. netlog-ai netlog-ai Public

    AI-powered network log analyzer with LLM-assisted root-cause analysis (Junos / EOS / FRR). Local Docker Model Runner or Anthropic Claude. Sanitize-before-LLM. 139 tests. MIT.

    Python 10 4

  4. claude-skill-lint claude-skill-lint Public

    Zero-dependency linter for Claude Code skills — frontmatter, description length, progressive disclosure, dead refs, and stale model IDs.

    Python 2 1

  5. argus argus Public

    Agentic AI penetration tester — reasons, chains, and adapts across network/host/AD/web, read-only by default behind a fail-closed 7-layer guardrail. Proof-annotated attack paths, HMAC-audited, Clau…

    Python 2

  6. aegis aegis Public

    Air-gapped pre-deployment network change validation against a real containerlab digital twin, with sealed PCI/SOC2/NIST evidence. Zero egress.

    Python 1 1