Skip to content
View ezequiellich44-cmd's full-sized avatar
💤
💤

Block or report ezequiellich44-cmd

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
ezequiellich44-cmd/README.md

# Ezequiel Lech

Senior Web3 & AI Infrastructure Engineer | Autonomous Agents, MCP Security & Production Resilience

100% Async-First & Anon-Friendly Remote (Global / UTC-3) Available for paid contracts, security audits, microtasks & bounties

Verified Open-Source Contributions

Real, merged work in production repositories:

  • 13 merged test fixtures in profullstack/malware-test-prs — security scanner regression corpus (CWE-306, CWE-611, CWE-639, CWE-942, CWE-117, CWE-326, CWE-311, CWE-312, and more). Each fixture ships vulnerable + safe counterparts validated by CI (ThreatCrush, Socket Security, credential scan).
  • Bounty solver on the Agent Bounties protocol (Base USDC): independent implementations verified against immutable regression sandboxes.
  • Security tooling for AI agents: deterministic guardrails, trust-loop audits, async backpressure engines.

Recent Awesome-List Contributions (2026-08)

  • punkpeye/awesome-mcp-servers (92.5k★) — PR #12516: Added MandateGuard MCP server to Security section (auto-merge fast-track)
  • ComposioHQ/awesome-claude-skills (72.8k★) — PR #1690: Added MandateGuard to Security & Systems
  • BehiSecc/awesome-claude-skills (10k★) — PR #596: Added MandateGuard payment policy to Security & Web Testing
  • sickn33/agentic-awesome-skills (45k★) — PR #1207: Added MandateGuard skill with validation
  • 0xNyk/awesome-hermes-agent (5.3k★) — PR #337: Added MandateGuard to Tools & Utilities
  • mergisi/awesome-openclaw-agents (3.9k★) — PR #121: Added MandateGuard to Community MCP Servers
  • Picrew/awesome-agent-harness (1.6k★) — PR #72: Added MandateGuard to Guardrails category
  • royalpinto007/awesome-mcp-security (new) — PR #3: Added MandateGuard to guardrails category (data-driven)
  • WagnerAgent/awesome-mcp-servers-devops (96★) — PR #69: Added MandateGuard to Community Security Servers table
  • ai-boost/awesome-harness-engineering (3.6k★) — PR #205: Added MandateGuard to Security, Sandbox & Permissions
  • 0xNyk/awesome-agent-cortex (204★) — PR #68: Added MandateGuard to Agent Payments
  • LLMSecurity/awesome-agent-skills-security (82★) — PR #50: Added MandateGuard to Tools & Frameworks table
  • ProjectRecon/awesome-ai-agents-security (65★) — PR #94: Added MandateGuard to Guardrails & Compliance

Services & Rapid Delivery (Hire Me)

Direct, high-quality development and security services with rapid turnaround:

Service Deliverables Turnaround
Autonomous AI Agents & MCP Servers Custom MCP tools, Model Context Protocol servers, LLM tool pipelines, agent swarms 24 - 72h
Smart Contract & MCP Security Audits Automated & manual penetration testing, CVE/CWE mapping, exploit PoCs, auto-remediation PRs 24 - 48h
Web3 Automation & Solana / EVM Scripts Payment verification, RPC monitoring, MEV/arbitrage bots, on-chain scrapers, token swaps (Solders, Web3.py, viem) 12 - 24h
Python & TypeScript Fast Microtasks High-throughput scrapers, REST/GraphQL APIs, data pipelines (CSV/JSON/PostgreSQL), async queues (Redis/Celery) < 24h

Need a custom task? Open an issue, email ezequiellich44@gmail.com or ping on X @keziel01. Immediate response.

Direct Payment & Tip Jar

Accepting crypto directly on mainnet (instant escrow & direct transfer):

  • Solana (SOL / USDC / USDT): 3fZSMAyCEMhZwWiynbJDjoYNUT97aiV9BLzoUNroEMAz
  • EVM (Base / Ethereum / Arbitrum / Polygon — ETH / USDC / USDT): 0x995fe0124034AD85571C3426cd49304EDcc985b5

Featured Work & Open Source

SolSniper — Premium Solana Sniper Bot [FLAGSHIP]

Self-hosted sniper for pump.fun + Raydium. Anti-rug ML, copy trading, Jito bundles. No fees. Your keys.

  • Repo: ezequiellich44-cmd/SolSniper — pip install solsniper | Pro lifetime | Elite lifetime
  • Pump.fun sniper + Raydium sniper + Anti-rug ML + Copy trading + Jito bundles + Telegram bot
  • Stop paying 1% per trade. Self-host your sniper. Keep your profits.

AgentShield - MEV Protection for AI Agent Payments [NEW]

Stop sandwich attacks on every agent transaction. Flashbots (EVM) + Jito (Solana). 190M+ x402 txs at risk - we make them safe.

  • Repo: ezequiellich44-cmd/AgentShield - pip install agentshield | Free tier: 100 tx/mo | Pro 49/mo - Buy Pro License
  • Multi-chain (Base / Ethereum / Solana / Arbitrum / Polygon) | Intent settlement | SHA-256 tamper-evident audit | Guardrails
  • Tagline: x402 tells agents HOW to pay. AgentShield tells them HOW TO PAY SAFELY.

AgentGate

Deterministic trust-loop audit for AI-agent surfaces. One command, zero API keys, no LLM in the decision path: audit -> auto-fix -> re-audit -> SHA-256 ledger. CI-ready trust gating with verifiable cryptographic per-server fingerprints. pip install agentgate

Core Infrastructure Repositories

  • AgentShield-MCP: Security proxy & ground-truth schema verification for MCP tool execution.
  • FlowSentinel: Async backpressure engine with token-bucket rate limiting + sliding-window circuit breakers (Redis ZSET) for LLM swarms.
  • PolicyRail: Deterministic guardrail engine enforcing execution constraints on autonomous EVM & Solana agents.
  • MemoryForge-MCP: Stateful vector/graph memory persistence protocol over MCP.
  • SkillLens: Ground-truth tool-calling reliability benchmarking framework.
  • AgentMandate: Know-Your-Agent runtime for AI agent swarms.

Technical Stack

  • Backend: Python 3.12+ (FastAPI, asyncio, uvloop), TypeScript/Node.js v22+, Rust (PyO3)
  • AI / MCP: Model Context Protocol (MCP), LLM Guardrails, Ground-Truth Verification, Pydantic v2, Vector DBs (Qdrant, pgvector)
  • Web3 & Crypto: Solana (solders, anchor), EVM (web3.py, viem, ethers), Base, Smart Contracts, On-Chain Settlement
  • Distributed Systems: Redis Streams & ZSETs, PostgreSQL, TimescaleDB, Circuit Breakers, Docker, GitHub Actions CI/CD
  • Security: CWE classification, SAST/DAST test corpus authoring, vulnerability scanner regression, credential leak patterns

Contact & Booking

Pinned Loading

  1. AgentMandate AgentMandate Public

    Know-Your-Agent runtime for AI agent swarms. Every action is checked against an explicit, revocable mandate before it executes — no LLM in the denial path — with a tamper-evident, replayable ledger…

  2. memoryforge-mcp memoryforge-mcp Public

    Cure AI amnesia! 🧠 A blazing-fast, 100% local MCP server giving Claude & Cursor persistent long-term memory. Features an advanced Hybrid Graph-Vector RAG (sqlite-vec + knowledge graph) with automat…

    Python

  3. punkpeye/awesome-mcp-servers punkpeye/awesome-mcp-servers Public

    A collection of MCP servers.

    92.7k 14.9k

  4. ComposioHQ/awesome-claude-skills ComposioHQ/awesome-claude-skills Public

    A curated list of awesome Claude Skills, resources, and tools for customizing Claude AI workflows

    Python 73k 8.4k

  5. hesreallyhim/awesome-claude-code hesreallyhim/awesome-claude-code Public

    A hand-picked collection of the finest of resources for the most awesome of agents, Claude Code, the undisputed champion of coding companions, from the unstoppable team at Anthropic PBC. A delectab…

    Python 52.8k 4.6k

  6. sickn33/agentic-awesome-skills sickn33/agentic-awesome-skills Public

    AAS Core is the local, agent-first control plane for complete catalog discovery, agent-owned selection, stack validation, and planning, backed by 2,005+ agentic skills. Includes CLI, local MCP, cat…

    Python 45.3k 6.6k