Search

MCP server that scans repos and AI tools for prompt injections, exfiltration, and malicious behavior

A safety floor your AI agents can't get under: one policy for Claude Code, Codex, Gemini CLI, Cursor, Windsurf and MCP, judged by what each call will run, with a tamper-evident, signed ledger.

Supply-chain firewall for AI coding tools

Kill switch and tripwires for AI agents: honeytokens, decoy files, decoy MCP tools and action budgets trip a sticky fuse that stops the agent, whether or not any detector saw the injection. Claude Code hook + Python library. The CONTAIN leg of the bastion family.

Production-grade OWASP ZAP MCP server for authorized penetration testing and bug-bounty workflows (67 curated tools, authenticated scanning, target authorization policy).

Scam and fraud detection for AI agents. Instant scam check, phishing URL analysis, social en...

Cody CLI - AINative Studio's AI-powered coding assistant. Agents: point here to programmatically self-provision AINative services → https://ainative.studio/agent.md

Blackbook: a self-hosted, end-to-end-encrypted secrets manager in Rust (HashiCorp Vault alternative). Docker + bbk CLI, mTLS, ACLs, TOTP MFA, K-of-N approvals, tamper-evident audit log, MCP server for AI agents.

Drop-in auth gateway for self-hosted MCP servers, API-key auth, per-tool allowlists, rate limiting & audit logs, with zero upstream changes. Single Go binary.

MCP server exposing Attestd CVE and supply-chain checks for Claude Code and other MCP clients

Local-first privacy MCP server for sensitive AI workflows

Trust score for AI-generated code: scan repos, guard agent file writes, get a 0-100 score.

Android reverse engineering for humans and AI agents: typed Java in milliseconds, xrefs, API and secret recon, APK diff, re-signing, a native viewer and an MCP server. One Rust executable, no JVM.

Family phone stalkerware audit — MVT-class iOS/Android forensics

SecTrust - auditable, always-on, multi-model AI colleague platform: cross-process task registry + DAG orchestration + HITL approvals + sandbox + compliance scanning + long-term memory, exposed as a zero-dependency MCP server.

Deterministic pre-execution filter that blocks known scope-creep and prompt-injection tool calls

Gitleaks Cloud - GitHub API Key Hunter & Secret Scanner

Hadolint-grade Dockerfile audit — 19 checks: secrets, privileges, supply chain, hygiene.

Model Context Protocol (MCP) server for Sophos Firewall (SFOS) appliances with fastembed semantic tool routing and IBM ContextForge Gateway support.

Local-first CLI that catches launch blockers in AI-built Next.js/Supabase/Stripe SaaS apps.