# Apply rule-based guardrails to agent traces and tool flows with Invariant > Insert a trace-aware guardrail layer between agents and their tools so unsafe message patterns or tool-call sequences are blocked by explicit rules. - Skill: `agentskillexchange/apply-rule-based-guardrails-to-agent-traces-and-tool-flows-w` (Agent Skill) - Install (CLI): `npx skillmds@latest add agentskillexchange/apply-rule-based-guardrails-to-agent-traces-and-tool-flows-w` - Raw SKILL.md: https://api.skillmd.com/api/skills/agentskillexchange/apply-rule-based-guardrails-to-agent-traces-and-tool-flows-w/raw - Safety review: pending (external: skill-scanner PASS, skillspector PASS) - Works with: Claude Code, Claude.ai, OpenAI Codex - Category: AI & ML - Author: agentskillexchange (https://skillmd.com/u/agentskillexchange) - Updated: 2026-09-08 - Page: https://skillmd.com/skills/agentskillexchange/apply-rule-based-guardrails-to-agent-traces-and-tool-flows-w --- # Apply rule-based guardrails to agent traces and tool flows with Invariant Insert a trace-aware guardrail layer between agents and their tools so unsafe message patterns or tool-call sequences are blocked by explicit rules. ## Prerequisites Python environment or Invariant Gateway deployment, target LLM or MCP-enabled agent workflow, guardrail rules or policies, sample traces or live requests to evaluate ## Installation Requirements and caveats from upstream: - Guardrailing rules are simple Python-inspired matching rules, that can be written to identify and prevent malicious agent behavior: - python - Here, (msg: Message) automatically is assigned every checkable message, whereas the second line executes like regular Python. To facilitate checking Guardrails comes with an extensive standard library of operations, a... Basic usage or getting-started notes: - Getting Started | - Source: https://github.com/invariantlabs-ai/invariant - Extracted from upstream docs: https://raw.githubusercontent.com/invariantlabs-ai/invariant/HEAD/README.md ## Documentation - https://invariantlabs-ai.github.io/docs/mcp-scan/guardrails-reference/ ## Source - [Agent Skill Exchange](https://agentskillexchange.com/skills/apply-rule-based-guardrails-to-agent-traces-and-tool-flows-with-invariant/)