Stop Claude Code from doing irreversible damage. Policy-gated execution + receipts so you can ship agents without sweating production.

ai-agents policy-enforcement llm-security auditability claude-code agent-governance deterministic-execution cryptographic-receipts runtime-guardrails

• Updated Feb 16, 2026
• Python

Your AI agent just burned 200ドル. AgentGuard stops it at 5ドル. Runtime cost guardrails for AI agents — budget enforcement, loop detection, kill switch. Zero dependencies, MIT licensed.

python tracing multi-agent openai developer-tools observability cost-control ai-safety ai-agents loop-detection guardrails llm langchain anthropic budget-guard ai-cost-management runtime-guardrails budget-enforcement

• Updated Feb 21, 2026
• Python