|
Clampd AI Agent Security: Tool Call Firewall for LLMs
Clampd is the runtime firewall for AI agents and MCP servers. 287 detection rules in microseconds, scope tokens, anomaly scoring. Self-hosted, open-core.
|
85 % |
|
Security guardrails for coding agents · agentjail
Offline policy guardrails for coding agents. Block dangerous tool calls, sandbox execution, and audit agent activity. Open source, Apache-2.0.
|
85 % |
|
node9 · Access control for AI agents
Access control for AI agents. You decide what your agents can do; node9 enforces it before they act: allow, review or block each tool call, with every decisi...
|
84 % |
|
Strathon: Open-source AI agent firewall
Write a CEL rule. Strathon blocks the tool call before it executes. 3 lines of code, 10 frameworks, self-host in minutes. The open-source AI agent firewall.
|
84 % |
|
ModelFuzz — Runtime Guardrails for AI Agents
ModelFuzz secures LLM agents against prompt injection. Scan for vulnerabilities, then block unsafe tool calls at the execution layer with one decorator.
|
84 % |
|
Airlock — Identity & Policy Enforcement for AI Agents
Cryptographic identity and deterministic policy enforcement for every AI agent tool call. Not guardrails — access control.
|
83 % |
|
AgentX: The Action Firewall for AI Agents
Block the catastrophic deterministically, coach the recoverable. AgentX is a runtime firewall for AI agents. It blocks catastrophic tool calls (DROP TABLE, s...
|
83 % |
|
Calus — Drop-in AI Security Gateway
Drop-in, detection-only AI-security gateway. Flags prompt injection, jailbreaks & agent abuse on every call. No code changes, never alters your traffic.
|
83 % |
|
ifivo: Agent Control Plane for AI (Policy, Approvals, Kill Switch)
ifivo is the runtime control plane for AI agents. Enforce spend limits, vendor allowlists, and risk policies. Approve high-risk actions from Slack or in-app....
|
82 % |
|
AI Agent Observability Platform | Guardonic
Monitor AI agent runs, prompts, tool calls, errors, latency, token usage, and cost with Guardonic's MIT-licensed SDK and hosted observability platform.
|
82 % |
|
hoop.dev | Runtime controls for AI agents
Data and context make agents useful. hoop.dev makes it safe. Attach a sidecar to a database, cluster, or agent sandbox and get masking, guardrails, and appro...
|
82 % |
|
GuardianAI — Your agent holds a wallet. One bad prompt shouldn't empty it.
Dual-layer security for autonomous AI agents: millisecond prompt-injection firewalls off-chain, verifiable ERC-8004 identity and tamper-evident evidence on-c...
|
82 % |
|
anytool - Infinite tools for your AI agents
One meta-tool for your agents. Generate, cache, and run any tool in isolated Bun containers with per-user keys and per-tool state—no new endpoints, no deploy...
|
82 % |
|
Trustabl AI | Make Your AI Agents Reliable and Production-Ready
Trustabl scans AI agent code for reliability, safety, guardrail, and tool-calling issues before they reach production.
|
82 % |
|
Clawsec - Open Source Security Guardrails for AI Agents
Your AI agent has root access. And zero guardrails. Clawsec is an open-source security plugin that blocks dangerous actions in under 5ms. One command: opencl...
|
82 % |
|
Clawdit — See What Your AI Agent Really Did
Open-source runtime security for AI agents. Intercept, audit, and enforce policies on every tool call. The black box recorder your agents need.
|
82 % |
|
Cog — Go-native AI agent runtime
Self-hosted AI agents in pure Go. Typed gears, hash-linked event log, byte-exact replay. Free tier ships; Pro unlocks chains and the machine-room control plane.
|
82 % |
|
GuardProxy — Read-Only API Proxies for AI Agents
Turn any full-access API key into a read-only key in 60 seconds. GuardProxy blocks destructive HTTP methods so AI agents can read data without modifying or d...
|
82 % |
|
AgentGuards — Guardrails for AI apps and coding agents
A drop-in guardrail layer that blocks prompt injection, jailbreaks, PII leaks, and data exfiltration before they reach your LLM. Simple subscription plans th...
|
82 % |
|
Repellox: Runtime Security for AI Agents in Production
Repellox stops prompt injections, jailbreaks, and unauthorized tool calls before they reach your AI agents. Runtime security for production.
|
82 % |
Detected social media networks