MandateGuard
Deterministic payment-policy and guardrail engine for autonomous AI agents, distributed with an MCP server. It evaluates budgets, tool scopes, destination allowlists, denylists, rate limits and Ed25519 signed mandates before money-moving tool calls execute, then records decisions in a tamper-evident SHA-256 chained ledger without putting an LLM in the decision path. Public GitHub repository created August 19, 2026.
MCP unverified
Integration
| Transport | stdio |
| Auth | none |
| Endpoint | https://github.com/ezequiellich44-cmd/MandateGuard |
| Install | python3 -m pip install git+https://github.com/ezequiellich44-cmd/MandateGuard.git[mcp] |
Use Cases
| 01 | Place a deterministic approval layer between an agent and tools or wallets that can move money |
| 02 | Enforce per-actor payment scopes, budgets, destination allowlists and rate limits before execution |
| 03 | Replay and audit payment authorization decisions from a tamper-evident local ledger |
Tags
agentic-commerce payments policy-engine guardrails mandates ed25519 audit-ledger python
Machine-readable: /api/servers.json
· JSON-LD schema embedded in <head>