Search

Local-first MCP router integrity auditor for AI agent workflows.

Build llms.txt from local Markdown or a sitemap.xml and check an existing llms.txt file.

Read-only queries on local phishing scan results: findings, IOCs, SPF/DKIM/DMARC.

Validate JSON against a schema with Ajv, infer a schema from samples and diff two schemas.

Local-first Base64, Luhn, key-component XOR, and synthetic ISO-0 PIN block tools.

Fairquery Agent Trust & Data Safety MCP server: PII/secret redaction, email validation, sanctions screening and dependency vulnerability checks for AI agents. Free local tools + optional paid Apify-backed deep tools.

Decode JWTs without verifying, flag risky algorithms and claims, verify HS256 or RS256 with a key.

MCP server for RepoFortify — structural production-risk scanning for AI-built repositories, exposed as Model Context Protocol tools for Claude Code, Cursor, and Windsurf.

Input sanitizer for AI agents — anti-prompt-injection shield

Dual LLM with Symbolic Memory: a reference-monitor middleware that neutralises indirect prompt injection and MCP tool poisoning by design (Python, mypy --strict).

An MCP server that gives AI agents deep semantic understanding of any production database.

AgentLock authorization middleware for MCP (Model Context Protocol) servers.

Session collection for AI coding agents — reads what agents write to disk and normalises it into one session model.

MCP server exposing Koma Gate prompt-injection classification as a tool for AI agents.

🛡️ BLA (Blind LLM Agents) — Adversarial defense shield that protects binaries, Electron apps, JS bundles, Python, and WebAssembly against AI reverse-engineering agents like REA. Poison -> Exhaust -> Hallucinate.

Query OSV.dev for known vulnerabilities by package and version, and scan lockfiles in batch.

User access reviews for AI agents and MCP servers, with audit evidence and OSCAL output

Open source runtime governance for AI agents.

Deterministic policy engine for autonomous agent tool enforcement

Test regular expressions in a timeout-guarded worker, explain them, find catastrophic backtracking.