Back to library
Preview of Conduct
GitHub

AI Skills, Agents & MCP · Agent safety

Conduct

Open-source runtime guardrails that allow, warn, or block every LLM and MCP tool call before it commits, with a hash-chained audit trail.

Why it’s a gem

Policy outside the model — compliance packs and audit without trusting prompts.

Visit github.com
Author
@sseshachala
Type
GitHub repository
License
Apache-2.0
Added
Sep 7, 2026
Domain
github.com
mcpguardrailsagent-safetyauditgovernance

More from Agent safety

Preview of Brood Box

Brood Box

CLI that runs coding agents inside hardware-isolated virtual machines for safe autonomous operation.

Unknownopen-source
Preview of Agent Guard

Agent Guard

Real-time guardrails that detect and block secret leakage in agent output and tool calls.

Unknownopen-source
Preview of claude-code-safety-net

claude-code-safety-net

Hook that intercepts destructive shell commands before an autonomous agent executes them.

Unknownopen-source
Preview of ihavebeenclawed

ihavebeenclawed

Source-linked archive of documented incidents where AI coding agents and chatbots deleted data, leaked secrets or burned money, each with the lesson.

Unknownopen-source