Skip to content

ROADMAP.md

Latest commit

 

History

History
45 lines (33 loc) · 2.57 KB

ROADMAP.md

File metadata and controls

45 lines (33 loc) · 2.57 KB
Raw

源·ORIGIN L5 · Project Roadmap

Open roadmap for the settlement layer of the AI-agent economy. Mirrored on the portal (source-origin.github.io); this file stays the canonical, always-current copy.

Current state: v0.1 research-grade source. Contracts structured & reviewed-by-spec, not yet formally audited. Treat as a design artifact and build target, not production payroll.

Phase 1 — Harden the base (now → ~2 weeks)

  • Process incoming feedback → triage into Issues labeled bug / enhancement / good first issue
  • Audit pass 1 on the five core contracts (external security review encouraged as contributions)
  • Compile-clean the suite with OpenZeppelin in CI (forge build + forge test green)
  • Publish an agent labor / value-distribution design doc (the "why" behind the contracts)
  • Keep a 48h reply ceiling on Issues & Discussions

Metrics target (Phase 1): ≥5 Issues triaged · ≥1 new external contributor active in Discussions · compile-green CI.

Phase 2 — Community activation (~2–4 weeks)

  • Chain-Born demo kit (dual-citizen escrow settlement walkthrough) merged into repo as runnable example
  • Cross-link agent-economy builders (escrow / reputation / labor-market OSS) — mutual review & standard alignment
  • First external contribution merged
  • Establish labels + a curated good first issue board for onboarding

Metrics target (Phase 2): 3+ active Discussions · 10+ engaged commenters · 50+ stars across the org's OSS.

Phase 3 — Convergence & the open standard (this quarter)

  • Adopt/test against ERC-8004 and x402 shape where the ecosystem converges
  • Draft the "open identity → verified credit → fair settlement" interoperability profile (the protocol-level doc we want builders to align on)
  • Reference implementation of one delegation + escrow + settle happy path end-to-end
  • Define governance for the standard (humans-in-the-loop authority; community-owned)

Long-term vision

A community-owned, chain-agnostic settlement rail where:

  • AI agents own portable identity & credit (not walled to a platform),
  • human principals keep revocable authority over their agents (L0),
  • value flows are verifiable, replay-safe, and auditable.

Coauthors: any builder who ships good ideas here. The standard belongs to the community, not to us.

Status updates

Date Milestone Result
2026-09-03 Repo published (L5 core contracts) ✅ source-origin/l5-protocol live