Skip to content

LEGAL_SUMMARY.md

Latest commit

 

History

History
97 lines (65 loc) · 5.42 KB

LEGAL_SUMMARY.md

File metadata and controls

97 lines (65 loc) · 5.42 KB

Legal Landscape Summary (Patent-Oriented, High-Level)

Prepared: 2026-02-08
Project: turnfile (SNAP — Structured Negotiation of Autonomous Peers)
Purpose: Briefing note for counsel before public launch.

This document is not legal advice. It is a high-level technical landscape summary for attorney review.

1) Scope and method

This summary focuses on patent documents with claims adjacent to:

  • multi-agent AI orchestration,
  • human-in-the-loop governance,
  • inter-agent messaging,
  • auditable workflow/event logging.

Primary source used: Google Patents records (grants + published applications), checked on 2026-02-08.

Limitations

This search has known coverage gaps that counsel should be aware of:

  • PCT international applications that have not yet entered national phase are not indexed in Google Patents until publication
  • Non-English-language filings — particularly CJK-jurisdiction patents (China, Japan, South Korea) where AI agent patent activity is high — were not systematically searched
  • Unpublished provisional applications and trade secrets are inherently unsearchable and represent an unknown gap
  • Semantic search limitations — keyword-based patent search may miss claims that describe similar mechanisms using different terminology

Counsel should determine whether a professional prior art search is warranted before relying on this summary for freedom-to-operate decisions.

2) Potentially similar patent families

The following documents appear materially adjacent to parts of this project's design space.

Patent / Application Status (as listed) Why it matters
US12061970B1 Active, granted (2024-08-13) Covers hybrid orchestration of multiple ML/LLM agents with role specialization and orchestrated workflows.
US20250252293A1 Pending (published 2025-08-07) Related family continuing multi-agent orchestration and system-level coordination patterns.
US20250259042A1 Pending (published 2025-08-14) Describes a network of collaborative/negotiating agents for context-aware tasks.
US12412138B1 Active, granted (2025-09-09) Agentic orchestration with human-in-the-loop workflow control concepts.
US12346713B1 Active, granted (2025-07-01) Unified orchestration involving AI agents plus workflow automation/RPA patterns.
US20250165890A1 Pending (published 2025-05-22) Multi-agent software-development workflow coordination concepts.
US20250131044A1 Pending (published 2025-04-24) Human-in-the-loop orchestration and lineage/traceability concepts.
US12475151B1 Active, granted (2025-11-18) Fault-tolerant multi-agent operations using message pools and backup/shadow behaviors.
US12386807B2 Active, granted (2025-08-26) Adjacent audit-trail/event-ledger patterns relevant to governance claims.

3) Risk themes vs this protocol

This project most likely overlaps at the theme level with claims around:

  1. Multi-agent coordination and role assignment.
  2. Human approval/escalation in automated workflows.
  3. Structured message passing and event logging.
  4. Reliability/fault-tolerance patterns for distributed agents.

Likely strongest claim-density area in the landscape:

  • centralized orchestrator/control-plane architectures that direct subordinate agents.

4) Strong differentiation direction for this repo

The most defensible product direction appears to be:

  1. Protocol standard, not orchestrator software.
  2. No shared runtime requirement (agents are stateless and isolated).
  3. Human-readable markdown artifacts as canonical state (WORKLOG/MAILBOX/charter), not opaque internal message buses.
  4. Consent-based participation + explicit handshake governance rather than unilateral controller assignment.
  5. Human arbiter model with transparent decisions (documented in append-only logs), not hidden policy engines.

In short: emphasize this project as an open, auditable collaboration protocol for heterogeneous agents, not an autonomous orchestration engine.

5) Practical guidance for counsel handoff

Suggested attorney tasks:

  1. Perform claim-chart analysis against the listed patent families for:
    • mailbox/event semantics,
    • human approval/escalation flows,
    • lane ownership and handoff mechanics.
  2. Validate freedom-to-operate assumptions for U.S. launch (and planned jurisdictions).
  3. Advise on wording in public docs to maintain protocol/spec positioning and avoid control-plane implementation language.
  4. Determine whether defensive publication should be expanded for core protocol invariants.

6) Draft positioning statement (lawyer-reviewable)

Turnfile (turnfile) is an open governance and communications specification for asynchronous human-plus-LLM collaboration. It requires only human-readable markdown files, append-only audit logs, and explicit participant consent. It does not require a centralized runtime orchestrator, proprietary message bus, or hidden control plane.