Zero-egress PII redaction for Claude. Runs locally — no data leaves your infrastructure.
Config is the same across clients — only the file and path differ.
{
"mcpServers": {
"io-github-edu963-ocultar-pii": {
"command": "<see-readme>",
"args": []
}
}
}Are you the author?
Add this badge to your README to show your security score and help users find safe servers.
Zero-egress PII refinery for AI pipelines. Runs in your infrastructure. Your data never leaves.
No automated test available for this server. Check the GitHub README for setup instructions.
Five weighted categories — click any category to see the underlying evidence.
No known CVEs.
No package registry to scan.
Be the first to review
Have you used this server?
Share your experience — it helps other developers decide.
Sign in to write a review.
Others in ai-ml / devops
Persistent memory using a knowledge graph
MCP server for using the GitLab API
Privacy-first. MCP is the protocol for tool access. We're the virtualization layer for context.
An open-source AI agent that brings the power of Gemini directly into your terminal.
MCP Security Weekly
Get CVE alerts and security updates for io.github.Edu963/ocultar-pii and similar servers.
Start a conversation
Ask a question, share a tip, or report an issue.
Sign in to join the discussion.
Zero-egress PII refinery for AI pipelines. Runs in your infrastructure. Your data never leaves.
[!IMPORTANT] Featured Article: OpenAI shipped a model. We built the system. 📖 OpenAI shipped a model. We built the system. Read on dev.to
| Stat | Value |
|---|---|
| SSRF bypass vectors found + fixed | 2 |
| Fail-closed scenarios tested | 6 |
| Vault persistence | Named Docker volume |
| Tier 2 engine | OpenAI Privacy Filter (Apache 2.0) |
| Key management | Doppler |
Welcome to the Unified OCULTAR Engine. This monorepo contains the core refinery, integrated applications, and enterprise security tiers.
/apps/ - Applications (Proxy, Sombra Gateway, SLM Engine, Dashboard, Automation Bridge, Web)/services/ - Core backend logic (Refinery, Vault, Mock API)/enterprise/ - Enterprise security extensions & licensing logic/internal/pii/ - Centralized PII detection engine & registry/extensions/ - Third-party AI tool integrations (Goose MCP, etc.)/docs/ - Technical and product documentation/security/ - Regulatory policies and integrity manifestsOCULTAR is built on a Zero-Trust for Data architecture. It is designed for senior security engineers who require verifiable guarantees before connecting internal data to external AI providers.
169.254.169.254 (IMDS) ranges with active DNS rebinding safety. 2 bypass vectors (including IPv6 loopback and non-standard decimal encoding) were identified and patched during adversarial testing.Tokenization is handled via a defense-in-depth pipeline that runs before any payload reaches an upstream AI provider.
| Tier | Shield | Technical Description |
|---|---|---|
| 0.1 | Base64 Evasion | Decodes, scans, and re-encodes PII hidden inside Base64/JWT blobs. |
| 0 | Dictionary | High-speed protection for VIPs, internal projects, and sensitive org names. |
| 0.5 | Pattern + Entropy | Shannon scoring for high-entropy strings, catching keys and tokens. |
| 1 | Rule Engine | EMAIL, SSN, IBAN (MOD97), CC (Luhn mod-10), 50+ national ID types. |
| 1.1 | Phone Shield | libphonenumb |