Claude Agent SDK Guardrails Kit
Tool policy, single-use approvals, spend caps and offline hook tests for the Claude Agent SDK (Python).
View Claude Agent SDK Guardrails Kit
Human-approved tool actions for LangGraph that survive a timeout or a crash without blindly repeating the write.

LangGraph re-runs a node from the top when you resume after an interrupt or a failure (documented replay behavior; LangGraph issue #9185). If a node sent a CRM update and then timed out, the replay sends it again. The pack's restart demo shows exactly this with the naive pattern: notes_in_crm: 2.
Python 3.10+. Pinned: langgraph 1.0.10, langgraph-checkpoint-sqlite 3.1.3, FastAPI. No API keys needed to run the tests or demos. 25 tests.
Zip, 28 KB, v1.0. Use in your own and client projects (one buyer: a person, or one company and its employees); no resale or giving the files away. Full terms in LICENSE.md.
Human-approved LangGraph tool calls that survive a timeout or crash without repeating the write.
Tool policy, single-use approvals, spend caps and offline hook tests for the Claude Agent SDK (Python).
View Claude Agent SDK Guardrails KitBooking backend for Retell AI and Vapi receptionists: availability, book, reschedule, cancel, transfer. Retry-safe, 44 tests (Python).
View AI Receptionist Booking BackendBuild with x402 payments: builder skills, a paid agent API starter, Hermes and ElizaOS plugin templates. 4 tested kits, $59 ($96 separately).
View x402 Developer Bundle