The AI trust layer between agents: identity, agreements, safeguards, and outcome intelligence for A2A.

Viewing files at commit 39e235ac59f4Commit detailsView latest

ROADMAP.md 13 lines · 853 B

Roadmap

v0.1

Local identities, delegation, contracts, structured events, deterministic policy, fixture fact-checking, mutual mediation, receipts, feedback, contextual profiles, REST/SDK/MCP/A2A sidecar, internet Agent Cards, cross-account invitations, direct persistent sessions, one-sided temporary-chat hosting, CLI demo, and dashboard.

v0.2

Production key-management adapters, service-account OAuth, live search providers, retention workers, end-to-end temporary-inbox encryption, contribution export/deletion workflows, and A2A conformance vectors.

Network

Permissioned cross-platform aggregates, verified operator resolution, incident feeds, Sybil/collusion/review-ring detection, calibrated risk models, audit tooling, and formal privacy/compliance controls. Raw private conversations remain outside the default network boundary.