Skip to content

Security: agentictrustlabs/agentic-primitives

Security

SECURITY.md

Security policy

What this repository is

A developer kit against testnet reference deployments (Base Sepolia). Sessions are demo-grade by design; production custody is the job of the KMS backends in @agenticprimitives/key-custody. Do not put real value through the reference rails, and do not report "the demo session is not production-grade" — that is documented, deliberate, and stated in the README.

Reporting a vulnerability

Do not open a public issue. Use GitHub's private vulnerability reporting on this repository (Security → Report a vulnerability), or email the maintainers listed on the GitHub organization.

Include: the affected surface (CLI, a wrapper package, the example app, a doc that induces an unsafe pattern), reproduction steps, and impact. Reports about the live rails or the published @agenticprimitives/* packages are forwarded upstream.

You can expect an acknowledgement within a few days. Please allow a reasonable disclosure window before publishing.

In scope

  • packages/create-primitives-app — anything the scaffold writes that weakens a generated project (a credential that reaches the browser, a widened issuer allowlist, a secret in a tracked file).
  • packages/home-connect, packages/interactions-client — verification logic, issuer policy, wire validation.
  • apps/commons — patterns the docs tell people to copy.
  • Docs that instruct an unsafe pattern.

The security model, in one paragraph

A token proves who and authorizes nothing; authority is an on-chain delegation the person signed, verified per call and revocable without the app's cooperation. Apps are delegates, never custodians — the worst case for a compromised app credential is a one-transaction revoke, never a lost identity. A report that shows a path from "compromised app" to anything larger than that is exactly what we want to hear about.

There aren't any published security advisories