Security & Monitoring MCP Servers
Security, identity, secrets, compliance, testing, logs, monitoring, alerts, and observability.
Bounded Tools Mcp
Published by GitHub account bounded-systems
Read-only MCP server that fetches bounded.tools' signed static JSON artifacts and verifies them against a Sigstore-signed manifest.
Tripwire
Published by GitHub account bonesdefi
Tripwire is a local MCP proxy that sits in front of your tool servers and blocks prompt-injection-driven tool calls before they execute.
Bomly
Published by GitHub account bomly-dev
Bomly exposes its dependency-graph CLI to coding agents: scan projects for dependencies and vulnerabilities, diff two targets, and explain why a package is…
Shield
Published by GitHub account bolyra
A local stdio proxy that wraps any MCP server to enforce per-tool permissions, replay protection and audit receipts on tool calls.
Sdk
Published by GitHub account bolyra
A local TypeScript SDK for Bolyra that gives an assistant building blocks for mutual zero-knowledge-proof authentication between humans and AI agents.
Bolyra
Published by GitHub account bolyra
A library that adds Bolyra zero-knowledge proof authentication to MCP servers, so only authorized agents can call sensitive tools.
Gateway
Published by GitHub account bolyra
A local reverse-proxy gateway that authenticates agents and enforces per-tool permissions before forwarding MCP tool calls to an upstream server.
Memwright
Published by GitHub account bolnet
A self-hosted memory layer for AI agents that stores, recalls and audits shared memories across SQLite, pgvector and Neo4j backends.
Attestor
Published by GitHub account bolnet
Self-hosted governed memory for multi-agent systems, with role-based access, provenance, bi-temporal replay, and audit logging.
Sovseal Memory
Published by GitHub account bobybarack
Gives an assistant a private, on-device semantic memory that stores and recalls context locally without sending plaintext to the cloud.
Hacker Bob public records
Published by GitHub account bobnetsec
Hacker Bob public records server: lets an assistant browse cleared public CVE records and, when authenticated, manage workspace assets, assessments, findings…
Agentguard47
Published by GitHub account bmdhodl
Read-only MCP access to AgentGuard coding-agent traces, alerts, costs, usage, and budget health.
Evidrift
Published by GitHub account bm1016bm-svg
Records TypeScript call signatures or repository JSON values as content-addressed receipts and revalidates them in CI to catch API contract drift.
Postgres Scout Mcp
Published by GitHub account bluwork
Lets an assistant inspect, diagnose, and optionally modify a PostgreSQL database through 38 tools for health, performance, and data quality.
Mongo Scout Mcp
Published by GitHub account bluwork
Lets an assistant explore, query, monitor, and diagnose MongoDB databases, with data-quality checks and optional guarded writes.
Agent Observability
Published by GitHub account blueskylineassets
Lets an assistant record structured logs, track costs, and keep compliance audit trails for agent activity.
Blueprint Agentic Staking (Solentic)
Published by GitHub account blueprint-infrastructure
Solana staking server for AI agents: stake, unstake and withdraw SOL with the Blueprint validator, plus read-only staking data and verification tools.
Model Ledger
Published by GitHub account block
Local MCP server for Model Ledger, a model inventory that lets an assistant trace deployed models, their dependency graph and change history.
Dear User
Published by GitHub account bleedmode
Audits a Claude Code or Claude Desktop setup locally, scoring collaboration quality and flagging leaked secrets, config conflicts, and broken structure.
Knowing
Published by GitHub account blackwell-systems
A local MCP server that indexes a code repository into a content-addressed graph and gives an assistant ranked code context, blast radius, test scope and verif-
BlackDome Threat Intel
Published by GitHub account blackdome-ai
Read-only honeypot threat intelligence: look up attacker IPs, browse IOCs, inspect credentials and payloads, and check ThreatDrop case status.
BlackDome Threat Intel
Published by GitHub account blackdome-ai
BlackDome Threat Intel gives an AI assistant read-only access to live honeypot threat intelligence: attacker IPs, IOCs, credentials, payloads and threat actors.
Conduit
Published by GitHub account bkauto3
A local headless browser server that gives an AI assistant browser control plus SHA-256 audit trails and Ed25519-signed proof bundles.
Diffpilot
Published by GitHub account bkalafat
A local MCP server that lets an assistant review your Git changes, draft PR titles, descriptions and commit messages, and scan diffs for secrets.