Awesome Hermes Plugins › Tools & Automation

intentframe/agent-integrations

⭐ 11 Python added to this list on 2026-07-15 repository created 2026-06-21

The intentframe/agent-integrations repository offers a security plugin that integrates IntentFrame, an external policy runtime, with Nous Research's Hermes Agent. This integration establishes a critical security layer by intercepting and validating high-risk actions initiated by Hermes Agent, such as terminal commands, code execution, file writes, and cron job scheduling, before they are executed on the host machine. Unlike Hermes's internal security features, IntentFrame operates as an 'outside' judge, ensuring that even a compromised or confused agent cannot bypass predefined security policies. The plugin routes selected Hermes tool calls through IntentFrame, where they are checked against user-defined policies. Governed tools, including `terminal`, `execute_code`, `write_file`, `patch`, and `cronjob`, are monitored by default. The project emphasizes a 'deny-by-default' policy model, requiring explicit allowances for actions. Installation is streamlined via a curl command, which also sets up the IntentFrame Control Plane for policy management, audit logging, and tool governance configuration through a web UI. This architecture enhances AI agent safety by providing an independent, auditable control plane for autonomous actions, preventing privilege escalation, unauthorized file access, and other malicious activities.

https://github.com/intentframe/agent-integrations

ai-agentai-agent-securityai-agentsai-governanceai-safetyai-securityai-security-governancehermeshermes-agentintentframellmllm-securityllm-security-prompt-injection-ai-safetynous-researchpolicy-as-codeprompt-injection-defensetool-gatingsecurityobservability