spend-control

The guardian layer for AI agents — identity, secrets, audit via MCP. Gate + Vault + Watch.

Agentic SpendGuard — audit-chain spend control for LLM agents. KMS-signed decisions, Stripe-style auth/capture ledger, operator approval, multi-tenant. Adapters for Pydantic-AI, LangChain, LangGraph, OpenAI Agents SDK, Microsoft AGT.

Spend control for Nigeria's informal micro-businesses — issue capped virtual cards to staff, freeze instantly, see spend live. Built on BMONI. Hackathon build for BMONI "Hack the Future."

Pre-flight spend authorization for AI agents. Reserve-confirm lifecycle, concurrency-safe, append-only audit ledger.

Vault your key, cap spend, cut cost, and get a signed receipt - for Cline, Kilo Code, OpenCode, and Zoo Code, across editor, CLI, and SDK.

Open-source AI gateway and spend-control proxy for LLM providers: per-key and global token budgets with hard-stop enforcement, tokens-to-cost accounting, and audit-ready usage. Rust, on Cloud Run.

Chat-native spend-control layer for AgentCash/x402 calls: wallets, quote ledger, policies, approvals, audit, and spend analytics.

Keep your AI agents on a leash: bind an agent's identity, give it a governed spending account with on-chain limits, and cut it off everywhere with one write. ENS + Hedera x402 + Privy.

TypeScript SDK for NORNR — spend governance for AI agents

AI Agent Context-Boundary Spend Gate: veto agent spends that cross context boundaries without re-authorization.

Download the App

Open-source, local-first deterministic spend-control gate for autonomous agents. Policy gate + hash-chained ledger + human approval. Verified, not narrated.

Give your AI agent a wallet it can't drain — gasless, policy-checked spend vaults on BOT Chain. Live: https://botspend-production.up.railway.app

Sourced registry of runaway-spend incidents caused by autonomous AI agents. Every record has a public source URL. CC BY 4.0.

Public discovery kit for the 400860 x402 API catalog

Distribution hub + open-data kit for sipi.bot — sourced AI-agent spending incidents registry. CC BY 4.0.

Per-task budget, loop detection, and kill-switch middleware for agent LLM calls. Deterministic, dependency-free, fail-closed.

TWZRD buyer-side x402 trust packages: x402 trust gate, elizaOS trust plugin, and the agent-intel client. Vet a counterparty before you pay.

Zero-dependency local proxy & fail-closed spend firewall to halt runaway AI agent token loops

The budget that actually blocks — per-agent spend guardrail for agentic payments. Reserve/commit/release with a proven concurrency guarantee, policy engine, and audit trail. x402-ready.