Lightweight Encrypted Agent Secret Handling — a standard for letting AI agents act on your behalf without ever holding your secrets.
AI agents are being handed credentials the way interns get handed the master key: fully, indefinitely, and with no record of what they did with it. A leaked agent config today can mean a leaked API key, a leaked password, a leaked life. The industry's current answer — paste your secret into the agent's environment and hope — is not an answer.
LEASH makes delegation a handoff, not a secret transfer:
The agent never sees the underlying credential. It receives a scoped, version-bound delegation that the vault can honor — and revoke.
Delegations name what the agent may do, for how long, and nothing more.
The person, not the platform and not the agent, is the root of authority. Every delegation is auditable by default.
LEASH is a standard, not a product. It works with agentic frameworks via a connector model (including MCP) without depending on any of them.
LEASH is being prepared for submission to the Agentic AI Foundation as a proposed vendor-neutral standard.
Read the LEASH specification on GitHub →
VettID will be one implementation of LEASH. The standard doesn't depend on VettID — that's the point.