I build security tools, then explain them to the people who have to buy them.
6 years working where cybersecurity, data and customer conversations meet: finding the real risk, designing the solution, proving it with a working demo and putting a number on the business case.
What I do
- Pre-sales engineering: discovery, solution design, demos and proposals with ROI, for technical and non-technical buyers
- Cybersecurity: vulnerability management, detection engineering, phishing analysis, attack-path modeling, compliance (NIS2, GDPR, ISO 27001)
- Data and AI: analytics, dashboards and AI used with guardrails in security workflows
Education: MBA in AI & Data Analytics (USP/Esalq), postgraduate in Cybersecurity Defense (FIAP), MBA in IT Leadership & People Management (Saskatchewan Polytechnic)
Languages: Portuguese (native), English and Spanish (fluent)
Every tool has tests, CI, a live demo, a SECURITY.md and, for the linters, --tsv output plus SARIF for GitHub code scanning. Several were validated on public corpora and the bugs that found are documented in each repo's docs/validation.md.
Start here: compliance-atlas, mcp-sentinel, attackpath, sigma-lint, presales-studio, gdpr-ropa-lint
Pre-sales and business case: presales-studio, discovery-copilot, questionnaire-copilot, breach-to-boardroom, risk-in-euros, fair-sim, insurance-readiness, external-posture, tlpt-planner, pqc-readiness, sovereignty-lens, compliance-workbench
Regulation and compliance (EU): compliance-atlas, nis2-gap-check, gdpr-ropa-lint, ai-act-inventory, dora-ict-register, cra-sbom-check, cra-readiness, ens-checker, cloud-to-ens, agent-inventory, risk-register, card-scan, cookie-audit
Detection and threat hunting: sigma-lint, sigma-lab, mitre-map, yara-lint, beacon-hunter, dns-exfil-hunter, auth-log-hunter, access-log-hunter, log-sentinel, logforge, evtx-triage, dfir-lens, incident-timeline, ioc-extract, ioc-hygiene, honeynet-live, canary-grid, terminal-soc, secops-hub
AI security: mcp-sentinel, mcp-red-team, soc-agent-bench, prompt-injection-arena, trifecta-guard, agent-inventory
Cloud, identity and infrastructure as code: iam-policy-lint, s3-policy-audit, cloud-iam-paths, azure-nsg-audit, entra-export-audit, ad-export-audit, attackpath, attack-path-prolog, rbac-audit, oauth-scope-audit, tf-guard, tfplan-guard, k8s-audit, k8s-netpol-audit, helm-values-audit, dockerfile-lint, dockerfile-audit
Supply chain and CI: gha-audit, jenkinsfile-audit, sbom-diff, sbom-license-gate, vex-reach, java-dep-audit, npm-install-audit, jar-scan, secret-proof
Hardening and host audits: linux-baseline, sshd-audit, ssh-config-audit, sudoers-audit, passwd-audit, login-defs-audit, systemd-unit-audit, iptables-audit, netconf-audit, nginx-audit, ini-audit, cron-audit, listen-audit, elf-harden, shell-risk
Application and web security: csp-audit, csp-builder, jwt-audit, php-vuln-lint, rails-sec-lint, wp-plugin-audit, redos-spotter, vulnshop, mail-auth-lint, mailauth-audit, dmarc-report, tls-expiry, cert-lint
Malware, phishing and forensics: phish-radar, phish-triage, pdf-triage, pcap-triage, packed-spotter
Vulnerability management and crypto: vuln-prioritizer, cvss31, crypto-lab, hash-audit, hashcheck-rs, passcheck-rs, entropy-c, portscan-go, surface-radar
Training and simulation: cyber-arena, tabletop-ir
En español: ens-checker, dora-ict-register, gdpr-ropa-lint, ai-act-inventory, nis2-gap-check, cra-sbom-check
| Project | Stack | What it does |
|---|---|---|
| compliance-atlas | Java | One control library mapped to NIS2, ISO 27001:2022, ENS, GDPR and DORA with an evidence-aware readiness score |
| presales-studio | TypeScript | Questionnaire to client-ready security proposal with ROI, EN/ES, number-validated AI summary |
| mcp-sentinel | Go | Policy proxy for MCP servers: blocks risky tool calls, redacts secrets, withholds injected results |
| soc-agent-bench | Python | Benchmark that scores SOC triage agents on verdict, evidence, actions and resistance to injection |
| tabletop-ir | TypeScript | Incident response tabletop with running NIS2/GDPR deadlines and a best-path after-action report |
| fair-sim | Python | FAIR-style cyber risk in euros: Monte Carlo loss curves, risk appetite and control ROI |
| secops-hub | Go | SOC dashboard with ATT&CK mapping and AI triage |
| prompt-injection-arena | TypeScript | Game and red-team harness for LLM prompt injection (OWASP LLM01) |
| attackpath | Go | Attack path graph with Dijkstra and remediation what-if |
| phish-radar | TypeScript | .eml analyzer: SPF/DKIM/DMARC, links, attachments, AI review that can only raise the verdict |
| breach-to-boardroom | TypeScript | Raw logs to a board-ready incident brief: euro loss range, NIS2 and GDPR clocks, decisions for leadership |
| cyber-arena | TypeScript | Deterministic red vs blue simulation with partial information, agent tournament and match logs |
| discovery-copilot | TypeScript | Pre-sales discovery prep: NIS2, DORA, ENS and GDPR scoping, ranked gaps, questions and demo storyline, EN/ES |
| compliance-workbench | TypeScript | RFP shredder mapped to NIS2, ISO 27001 and ENS, DORA contract register check, ENS planner, security ROI lab and a NIS2 CI gate |
| mcp-red-team | Python | Offensive tester for MCP servers: tool poisoning, injection, SSRF, secret leaks and rug pulls |
| cra-readiness | TypeScript | EU Cyber Resilience Act readiness: SBOM quality, obligations by product class and the 24h, 72h and 14-day reporting clock |
| external-posture | TypeScript | External attack surface posture from DNS, mail authentication, TLS and headers, with offline snapshots |
| agent-inventory | TypeScript | Inventory and risk classification of AI agents and MCP servers, with EU AI Act exposure |
| sigma-lab | TypeScript | Detection-as-code lab: Sigma precision and recall on labelled events and ATT&CK coverage of your threat profile |
| phish-triage | TypeScript | Offline phishing e-mail triage with an explainable verdict |
| tlpt-planner | TypeScript | DORA TLPT planner: dated schedule from the RTS 2025/1190 phases, readiness and cycle checks |
| cloud-to-ens | TypeScript | Cloud configuration to ENS and NIS2 findings, in Spanish and English |
| insurance-readiness | TypeScript | Cyber insurance questionnaire answers checked against evidence, with gates and misstatement risk |
| trifecta-guard | TypeScript | Replays agent traces against the lethal trifecta (private data, untrusted input, external communication) and suggests a policy |
| questionnaire-copilot | TypeScript | Draft security questionnaire answers from your own policies, with citations, confidence and stale-source flags |
| pqc-readiness | TypeScript | Post-quantum readiness: crypto inventory, harvest-now risk against EU and NIST dates, migration waves and CBOM |
| sovereignty-lens | TypeScript | Cloud sovereignty screening: indicative SEAL per objective, GDPR Chapter V, DORA exit and sovereignty-washing flags |
| beacon-hunter | Rust | Finds C2 beaconing in connection logs with jitter-tolerant statistics and an explainable score |
| secret-proof | Go | Offline secret triage: separates real-format tokens from fakes via checksum verification, with JWT analysis |
| risk-in-euros | Python | Cyber risk in euros: FAIR-style Monte Carlo, loss exceedance curve and control ROI for budget decisions |
| dns-exfil-hunter | Python | Hunts DNS tunneling and DGA activity in query logs with explainable per-client scoring |
| gha-audit | Ruby | GitHub Actions workflow auditor: script injection, pull_request_target, unpinned actions and token permissions |
| auth-log-hunter | Perl | SSH auth log hunter: brute force, password spraying and compromise detection from auth.log |
| sbom-license-gate | Java | Licence policy gate for CycloneDX SBOMs: SPDX expressions, copyleft review and CI exit codes |
| pcap-triage | C | Packet capture triage: port scans, host sweeps and cleartext credentials from pcap files, fuzz-tested |
| packed-spotter | C++ | Packed and encrypted binary spotter: entropy map, PE section analysis and a 0-100 score, fuzz-tested |
| wp-plugin-audit | PHP | Static security audit for WordPress plugins with light taint tracking |
| access-log-hunter | Lua | Web attack hunter for access logs: SQLi, traversal, log4shell and scanners with per-IP scoring |
| cron-audit | Nim | Persistence hunter for crontab files: reverse shells, download-and-run and hidden payloads |
| iam-policy-lint | OCaml | AWS IAM policy linter: wildcards, public principals and privilege escalation paths |
| ssh-config-audit | Elixir | sshd_config hardening audit with an A to F grade and the exact fix for each line |
| ad-export-audit | C# | Active Directory export audit: Kerberoasting, AS-REP roasting, delegation and stale privileged accounts |
| mail-auth-lint | Haskell | SPF, DMARC and DKIM linter: finds the small silent mistakes that leave a domain spoofable |
| vuln-prioritizer | R | Risk-based vulnerability prioritization: CVSS, EPSS, KEV and exposure into P0 to P3 tiers |
| cert-lint | Erlang | Offline X.509 certificate linter: expiry, weak keys, SHA-1, SAN and quantum exposure |
| cvss31 | Fortran | CVSS v3.1 base score calculator and batch triage, spec-exact rounding |
| attack-path-prolog | Prolog | Attack paths from facts: internet to crown jewels, with choke points |
| card-scan | Ada | Finds card numbers in files for PCI DSS scoping, always masked |
| redos-spotter | Common Lisp | Finds ReDoS-prone regexes and proves it with step counts |
| dmarc-report | Racket | DMARC aggregate report analyzer: spoofing sources and enforcement readiness |
| netconf-audit | Tcl | Cisco IOS style config audit: telnet, SNMP, secrets, VTY exposure |
| listen-audit | AWK | Audits ss/netstat output for risky listening services (AWK) |
| hash-audit | D | Password hash file audit: weak algorithms, empty passwords, reuse |
| jwt-audit | Rust | JWT audit: alg none, key injection, expiry and secrets in claims |
| tfplan-guard | Go | Terraform plan review before apply: data loss, open ports, wildcard IAM |
| oauth-scope-audit | Python | OAuth app grant audit: risky scopes, unverified publishers, stale access |
| jar-scan | Java | Java archive scanner: vulnerable libraries inside JAR, WAR, EAR and shaded jars |
| elf-harden | C | ELF hardening checker: PIE, NX, RELRO, canary, RPATH, unsafe imports |
| pdf-triage | C++ | PDF triage: JavaScript, launch actions, embedded files, obfuscated names |
| nginx-audit | Ruby | nginx config audit: weak TLS, alias traversal, missing headers |
| ini-audit | PHP | php.ini audit: remote includes, error display, session cookies, dangerous functions |
| csp-audit | Scala | HTTP security headers and CSP audit: unsafe-inline, wildcards, HSTS, framing |
| sudoers-audit | Clojure | sudoers audit: NOPASSWD ALL, shell escapes, wildcards, editors that give root |
| passwd-audit | COBOL | passwd and shadow audit in COBOL: UID 0, empty passwords, weak hashes |
| iptables-audit | Pascal | iptables audit in Free Pascal: ACCEPT policies, open admin ports, missing drops |
| login-defs-audit | Guile | login.defs audit in Guile Scheme: password aging, hash method, umask |
| rbac-audit | Crystal | Kubernetes RBAC audit: cluster-admin, wildcards, secret readers, exec, escalate |
| systemd-unit-audit | Vala | systemd unit hardening audit: root, sandboxing, capabilities, secrets |
| npm-install-audit | JavaScript | npm install-time risk audit: install scripts, git deps, wildcards, HTTP tarballs |
| nis2-gap-check | Raku | NIS2 Article 21 gap check from a CSV self-assessment: missing, unowned, unevidenced and stale measures |
| dora-ict-register | Standard ML | DORA register of information checker: missing fields, LEI and contract date errors, critical function gaps |
| ens-checker | Scheme | Spanish ENS compliance checker for facts files: category, measures and evidence gaps |
| cra-sbom-check | F# | Cyber Resilience Act SBOM check: missing supplier, version, hash, licence and vulnerability data |
| ai-act-inventory | Python | EU AI Act inventory linter: risk class, role, oversight and documentation gaps per AI system |
| gdpr-ropa-lint | Go | GDPR Article 30 record of processing linter: missing purposes, legal bases, retention and transfer safeguards |
| s3-policy-audit | Rust | AWS S3 bucket policy and configuration audit: public principals, wildcards, TLS, encryption, logging |
| azure-nsg-audit | Visual Basic .NET | Azure NSG export audit: internet-facing admin and database ports, any-any allows, wide ranges, egress |
| entra-export-audit | C# | Entra ID export audit: admins without MFA, permanent roles, guests with admin roles, stale accounts |
| k8s-netpol-audit | Nim | Kubernetes NetworkPolicy coverage audit: unprotected namespaces, default deny, allow-all rules |
| helm-values-audit | Lua | Helm values.yaml audit: hard-coded secrets, weak passwords, privileged containers, latest tags |
| sigma-lint | Ruby | Sigma rule linter: required fields, UUIDs, conditions, unused selections, match-all values, ATT&CK tags |
| yara-lint | Haskell | YARA rule linter: missing conditions, unreferenced strings, short strings, slow regexes, duplicate names |
| ioc-hygiene | Perl | IOC list linter: live URLs and IPs, reserved addresses, known-good indicators, bad hashes, duplicates |
| evtx-triage | D | Windows event log triage from JSON: log clearing, RDP, brute force, services, PowerShell, LSASS |
| mitre-map | PHP | ATT&CK reference checker: malformed and deprecated technique IDs, redundant tags, tactic coverage gaps |
More tools (32 smaller repos)
Practical tools that connect security, operations and business value. Each repository has tests, CI and a README that explains the design decisions.
| Project | What it does | Stack |
|---|---|---|
| honeynet-live | Live honeypot with attacker simulator, SSE dashboard and IOC export | Go |
| crypto-lab | SHA-256 and AES from scratch plus ECB, padding oracle, length extension and RSA attacks | Rust |
| terminal-soc | Terminal SOC console (ratatui) with alert triage and playbooks | Rust |
| vulnshop | Deliberately vulnerable shop with a secure twin and a scanner that proves the difference | PHP + Go |
| logforge | Freestanding log analysis core with CLI, Python and Node bindings, benchmarked | C |
| vex-reach | Advisories + lockfile + source into an OpenVEX document, using import reachability | Rust |
| cloud-iam-paths | Privilege escalation paths to admin in an AWS IAM export, with the shortest chain | Go |
| canary-grid | Signed canary tokens and decoys with a hardened listener and a log scanner | Rust |
| dfir-lens | Local-first forensic timeline from Windows, Linux and web logs with ATT&CK-mapped findings | TypeScript |
| surface-radar | Attack surface report and change diff from nmap XML, never touches the network | Java |
| vulnprio | Risk-based vulnerability prioritization with CVSS, EPSS, CISA KEV and asset context | Python |
| detect-engine | Sigma-style detection engine with MITRE ATT&CK mapping and Navigator export | Python, YAML |
| compliance-forge | NIS2 + GDPR + ISO 27001 gap assessment with a phased remediation roadmap | Python, YAML |
| presales-roi | ROI, NPV, IRR and sensitivity analysis for business cases | Python |
| policy-rag | Grounded Q&A over security policies with citations and retrieval evaluation | Python, BM25 |
| hardening-audit | Read-only Linux audit that generates an Ansible remediation playbook | Python, Ansible |
| agentshield | Security scanner for MCP servers, agent skills and prompt files: local web dashboard, SARIF and GitHub Action | Python, FastAPI, JavaScript |
| actions-audit | Security linter for GitHub Actions workflows: unpinned actions, script injection, risky permissions | Python |
| secret-sniffer | Secret scanner with entropy detection, staged-commit scanning, baselines and redacted output | Python |
| security-headers-audit | A-F grading of HTTP security headers, CSP and cookie flags | Python |
| jwt-inspect | Offline JWT auditor: alg none, weak HMAC secrets, jku/kid abuse | Python |
| portscan-go | Concurrent TCP scanner with banner grabbing and exposure labels | Go |
| hashcheck-rs | File integrity manifests, verification and duplicate finder | Rust |
| csp-builder | Build, parse and lint Content-Security-Policy headers | TypeScript |
| java-dep-audit | Flags known-vulnerable Maven and Gradle dependencies, XXE-safe parsing | Java |
| entropy-c | Shannon entropy analyzer to spot packed or encrypted files | C |
| rails-sec-lint | Static security linter for Ruby and Rails, 16 rules with CWE ids | Ruby |
| php-vuln-lint | Static linter for insecure PHP patterns: SQLi, XSS, LFI, RCE | PHP |
| linux-baseline | Read-only Linux hardening checker with JSON output | Bash |
| tls-expiry | TLS certificate checker: expiry, chain, hostname, weak crypto, many hosts at once | Go |
| log-sentinel | SSH log analysis: brute force, spraying and success-after-failure detection | Go |
| mailauth-audit |