Soki (想起)

Note: This analysis is based solely on the self-reported project description provided by the author. No external verification, historical data, or third-party sources are available. All claims are treated as stated by the author, not proven facts.

The description states that Soki is a decision-aware memory prototype for long-term human–AI collaboration. It is built as a small Node.js web application with a deterministic decision-resolution core, and includes:

• An 8-stage resolver process:
  • Validate the memory graph
  • Retrieve relevant candidates
  • Expand relationships
  • Apply state rules
  • Suppress invalid or superseded memories
  • Resolve authority and conflicts
  • Build a bounded decision packet
  • Produce final answer and reference explanation

• A deterministic core that decides which memories may be used, suppressed, or blocked.
• An optional GPT-5.6 Responses API boundary for presentation and explanation — but not for overriding decisions.
• Correction events stored as append-only JSONL files with before-and-after snapshots and reasons.

The product is described as a prototype, not a production-ready system.

The author positions Soki as a solution to a specific problem in long-term human–AI collaboration: AI systems failing not because they forget information, but because they fail to preserve decision state. This includes:

• Current vs. obsolete information
• Superseded or rejected alternatives
• Source authority
• Rationale behind decisions

The project evolved from a personal challenge into a prototype that demonstrates three use cases:

1. Current price resolution
2. Author-led literary revision
3. Rejected-candidate recovery

The positioning is framed around preserving human decision-making authority, not replacing it.

The description does not name specific customer segments or personas. However, the author implies that Soki targets:

• Users working on long-running creative or publishing projects
• Teams managing complex, evolving decisions in product development, research, or operations
• Individuals or organizations using personal knowledge management systems (e.g., Obsidian)

The ICP is inferred from use cases described but not explicitly defined.

There is no evidence of a business model or pricing structure. The description states that Soki is a prototype, and no revenue, customer data, or monetization strategy are mentioned.

Not evidenced.

The system is built using:

• Node.js
• JavaScript
• HTML5, CSS3
• JSON, JSONL
• OpenAI APIs (GPT-5.6 Responses)
• GitHub for version control

Key technical features include:

• Deterministic decision core
• Append-only event storage with replay capability
• Eight-stage resolver process
• Separation of concerns between decision logic and presentation layer
• 58 automated unit and integration tests
• UI that shows selected, suppressed, and conflicting memories

The system is described as fully functional in deterministic mode without API keys.

The description states:

• 58 automated tests passed with zero failures
• All three demos verified through browser UI
• No purchaser data or customer names included in demo data
• Codex contributed to development and review
• Git history shows contribution from pre-Codex baseline to final implementation

No evidence of real-world usage, customers, revenue, or adoption beyond the author’s own testing.

The description does not mention any competitors. It is unclear whether Soki is positioned against existing tools for knowledge management, decision tracking, or AI collaboration platforms.

Not evidenced.

• Prototype only: No production-grade features, storage, or scalability indicated.
• No external validation: The system has not been tested in real-world settings or with users beyond the author.
• Self-reported maturity: Claims of 58 tests and deterministic behavior are unverified without independent audit.
• Unclear commercial viability: No pricing, monetization, or customer acquisition strategy described.
• Limited team size: Only one member (Miki Kamino) is listed.

These points raise concerns about readiness for market entry or investment.

1. What specific use cases have you tested Soki with beyond the three demos?
2. Have you conducted any user studies or feedback sessions with actual users?
3. How do you plan to scale beyond deterministic mode and integrate with existing tools like Obsidian or Notion?
4. What is your roadmap for production-grade features such as event storage, signatures, access control, and migration?
5. Are there any plans to monetize Soki or offer it as a service?