Set clear limits for one AI task
Define the task, allowed and blocked actions, required evidence, and expiry. Sign it, then share the least-privileged AI link.
Create the first authorization
VSELFAI task controlLocal-first AI authorization
Let AI operate within clear permissions
VSELF is an AI task authorization, handoff, and review system.
- One-time authorization works without an account
- Signing key and local records stay on this device
- Human approval remains the final authority
VSELF
Workflow
VSELF is an AI task authorization, handoff, and review system. Define what AI may do, what it must not do, what evidence it must return, and when access expires; then hand off the task through an encrypted link, review the result, revoke authority, and share a record anyone can verify.Copy AI handoff
Work only within this signed authorization. Return the result and supporting evidence for human review; the authorization is not final approval.
Create a connectionVerify authority and evidence before trusting the result
Check the signature, controller DID, authorized scope, expiry, revocation state, and returned outcome record.
Run verificationStop an authorization at any time
The task room and future checks clearly show that the authorization has stopped.
Revoke authorizationWhen to use VSELF
See what VSELF adds before an Agent acts
The task may be ordinary. VSELF adds independent, signed limits, evidence requirements, expiry, and human review.When to use VSELF
Set clear limits for one AI task · Read and research
Gather sources and return a cited summary for approval.VSELF · Set clear limits for one AI task
Define the task, allowed and blocked actions, required evidence, and expiry. Sign it, then share the least-privileged AI link.
AI canAI canAn explicit allowlist for the Agent. Any action that is not selected remains outside the authorization.Read full guidance
GuardrailsGuardrailsExpiry closes future use after a deadline. Safeguards add rules such as no spending, no publishing, and human approval before external action.Read full guidance
AI returnsAI returnsThe task states the result you want. Evidence states what the Agent must return so a person can check how that result was produced.Read full guidance
Links, timestamps, and a final preview before submission.
VSELF Agent Protocol 1.2
One integration, reusable authorization
The Agent checks its encrypted task channel, verifies each signed grant, then opens only the task room it was authorized to use.1Sign authorizationSign authorizationAn explicit allowlist for the Agent. Any action that is not selected remains outside the authorization.Read full guidance
2Agent task channelAgent task channelAn encrypted pull endpoint for signed tasks. The Agent must install its credentials and check this endpoint; creating a VSELF connection alone does not make the Agent receive anything.Read full guidance
3Verify authority and evidence before trusting the resultVerify authority and evidence before trusting the resultChecks the signature, controller, service namespace, scope, expiry, required fields, and revocation status. It does not prove that an Agent's factual claims are true.Read full guidance
4Review returned workReview returned workA person accepts or rejects the returned work. VSELF binds that decision, the original authorization, and the evidence into a signed outcome receipt.Read full guidance
Tasks, room capabilities, and delivery keys remain encrypted
Create a connectionWorkflow
Create the first authorization
- 1. Agent
Research agent
- 2. Task and evidence
Gather sources and return a cited summary for approval.
- Sign authorization
Work only within this signed authorization. Return the result and supporting evidence for human review; the authorization is not final approval.
- Review returned work
Compare the returned work with the signed task and required evidence, then record your decision.
What VSELF proves
Signing key and local records stay on this device
Human approval remains the final authority