SafeAgent

Durable execution-claim guard for AI agents and SaaS applications.

Suppresses repeated logical actions and keeps uncertain outcomes visibly PENDING. Claim before execution, settle after a receipt, and reconcile unresolved attempts with the external provider.

State machine

CLAIMABLE → PENDING → COMMITTED; later claims observe PENDING or SKIP.

Endpoints

MethodPathDescriptionCost
POST/claimGate an action — returns PROCEED, SKIP, or PENDING$0.001 USDC
POST/claim/testFree test endpoint (10 calls/IP)Free
POST/settle/{id}Commit a PENDING claimFree
GET/auditFull claim history with filters$0.005 USDC
GET/healthLiveness probeFree

Quick start

curl -s -X POST https://safeagent-production.up.railway.app/claim/test   -H "Content-Type: application/json"   -d '{"agent_id":"my-agent","action_type":"send_payment","scope":"customer:123"}'
# First call: {"status":"PROCEED","test":true,"calls_remaining":9}
# Retry:      {"status":"SKIP","test":true}

GitHub · API Docs · PyPI