- Conduid
- Marketplace
- #agent-safety
MCP servers tagged agent-safety
92 live MCP servers tagged "agent-safety", ranked by trust score. Tags come from package metadata and repository topics, so this list covers servers across every category that work with agent-safety.
Dcl Skills
Security & compliance skills for AI agents (jailbreak defense, PII/secret redaction, hallucination checks) with a DeFi-specialized suite — free instruction-only mode or paid on-ch…
Undo MCP
MCP server for reversibility intelligence — check if actions can be undone
Constitution Lint Action
GitHub Action that lints CLAUDE.md-style agent constitution files for missing operational guardrails — spend limits, injection defense, escalation paths, secrets rules
ScrubMyText Examples
Runnable REST and MCP examples for deterministic AI-agent text safety, idempotency, webhooks, and human approval.
Relay Action Card
Run a Neura Relay Action Card and receive a governed Decision Receipt before execution.
Occp Core
OCCP — OpenCloud Control Plane: Governance-first AI agent orchestration with Verified Autonomy Pipeline. Policy-gated, audit-logged, kill-switch protected. 14 MCP tools, 8 special…
Howdex
The audit and verification layer for AI agents - verified procedural memory with compliance reports (SOC 2, EU AI Act, NIST AI RMF), a public procedure registry, and the verifier…
Qedra
A runtime gate for coding agents: an MCP server that blocks the tool calls that wreck a repo (force-push main, rm -rf, secret exfil, CI wipe) and lets real work through. Machine-c…
Prompt Injection Defense
The language-level attack defense skill every agent should keep. Covers 12 attack categories. Works with Claude, GPT, Gemini, Copilot, and any LLM.
Verification Security
Public verification, security review, and skill quality process for Agent Skill Exchange
Maqam
Give AI agents hands without handing them the keys: exact one-use approvals and verifiable receipts for registered actions.
Dualkey Agent
Approvals, deterministic policy, and signed receipts for AI agent actions.
Gate.cat
Block irreversible AI-agent actions before they run — deterministic, fail-closed action veto for Claude Code, Codex, Cursor, and other tool-using agents.
Rapid Agent Guardian
Agent safety gate for real work: evidence envelopes, human approval checkpoints and release guardrails for AI app builders.
Command Scope Contract
Bounded shell and CLI execution for AI agents: structured contracts, policy-gated execution, hardened Linux runtime enforcement, and signed receipts.
Munch Food Agent
A food-ordering agent that can't order food you didn't approve. Guardrails between the LLM and the world: hard dietary/allergen blocks, spend caps, and a mandatory human confirmat…
Mini Erp
Safe AI operations agent demonstrating MCP tools, deterministic ERP workflows, approval gates, evaluations, and idempotent writes.
Fairway
Tells you what to do in Claude Code when you don't know what to ask, plus hooks that encourage patterns and stop antipatterns. A monolith of small devtools sharing one Go binary a…
Policy Guard
Policy and safety middleware for MCP servers — authorization, rate limits, confirmation gates, audit logging, and PII redaction.
Cycles Claude Plugin
Hard budget enforcement for Claude Code tool calls — PreToolUse gating, AI-agent cost control, Cycles usage metering, and MCP budget tools.
Synartesis
An undo layer for AI agents. Records what an agent changed through MCP, puts it back, and refuses to let it do the unforgivable unasked.
Claude Code Governance
Runtime governance framework for Claude Code AI agents. Enforce boundaries, validate intent, and structure autonomous workflows — through code, not prompts.
Scalebotics Sable Test Agent Fixture
Controlled public fixture repo for repeatable Sable scan-to-policy-to-runtime validation demos.
Totp Presence
TOTP-based human presence signal for AI agents
Housewarden
Household-operations MCP server: every mutating tool is dry-run, confirmed and audited. Streamable HTTP, MCP spec 2026-07-28. Built for Alexa+.
Triple Arbiter
x402 settlement facilitator + EAS-compatible threat-intel attestation issuer on Base mainnet
Sqlgate
A deterministic, fail-closed write gate for LLM-generated SQL, exposed over MCP. The model can't talk its way past it.
Tollgate
Safety layer for AI agents — Protect · Route · Prove. Budgets, tool-loop hard stops, failover, chaos DR proof. OpenAI drop-in · MCP · n8n.
Winops Agent
An MCP agent that diagnoses Windows service faults through least-privilege PowerShell JEA endpoints and cannot change anything without human approval.
Deployment Patterns
Deployment patterns and operational best practices for running MCP servers safely in production.
Cycles Client Python
Python SDK for Cycles — reserve, commit, release budget and risk for AI agents
Shutter Cull MCP
The write-capable half of shutter-cull for agents: it can propose a cull, it cannot execute one without an exact confirmation phrase typed back, and every write it makes can be un…
Agentiam Recipes
Short, opinionated integration recipes for AgentIAM — 18 x402 endpoints on Base Mainnet for agent identity, safety & execution proofs.
Contract Review MCP
Read-only MCP server patterns and examples for contract review workflows.
Legal Toolkit
MCP tools and examples for safe, local-first legal and compliance workflows.
Nightcap
🌙 A hard stop for agents you leave running: dollar caps, pinned models and a crash-safe action ledger for MCP tools and Claude Code
Deterministic Safety Sandboxes With MCP
Dependency-free Python MCP server that puts a deterministic policy boundary in front of subprocess execution for agents.
Seal Live Demo
Browser-runnable demo of the seal verified Policy Decision Point: replayed tool-call attacks against the live deterministic WASM kernel.
Awesome Agent Safety
Awesome Agent Safety: curated agent-specific safety - threat models, guardrails, MCP/A2A security, evals, incidents, governance
Think MCP
MCP server for intent security pre-flight checks for autonomous AI agents
Agit
Undo, history, and proof for AI agents.
Kinesift Public
Observe AI agents from outside the sandbox: tamper-evident side-effect logs for Docker-run agents, with Harbor import and reset-able task environments.