--- mw_bundle: 1 id: 8IkpnVgX title: "Engineering decisions + cross-AI strategy" url: https://memory.wiki/b/8IkpnVgX document_count: 3 updated: 2026-05-14T19:20:02.321Z analysis_generated_at: 2026-05-14T19:20:02.321Z source: "memory.wiki" --- # Engineering decisions + cross-AI strategy > A working bundle: the ADR for shipping graph_data in URLs, the cross-AI moat argument, and a hands-on integration note for Cursor. Three docs that explain *what we're building and why* in 10 minutes of reading. **Intent:** Frame the platform decisions a reviewer would want to verify in 10 minutes. ## Summary These documents establish mdfy's platform strategy centered on cross-AI portability through bundle URLs that embed rich analysis data. The technical decision to ship graph_data inside URLs enables a viral loop where recipients get more value than senders could provide elsewhere, creating a structural competitive moat based on vendor neutrality that individual AI companies cannot replicate. ## Themes - cross-platform portability - competitive differentiation - technical architecture - integration workflows ## Cross-document insights - The decision to embed graph_data in URLs transforms a simple sharing mechanism into a competitive moat - no single AI vendor can replicate cross-competitor portability without undermining their own platform strategy. - The viral loop mechanism depends on URLs providing more value to recipients than senders could deliver through native platform features, making the technical implementation decision strategically critical. - Platform integration patterns reveal that mdfy succeeds by fitting into existing workflows rather than requiring users to adopt new tools, with different surfaces (Cursor, ChatGPT, Claude) needing different integration approaches. - The tension between token cost optimization and strategic value delivery suggests that technical trade-offs must be evaluated through the lens of competitive positioning, not just operational efficiency. ## Key takeaways - Cross-AI portability through bundle URLs creates a structural competitive advantage that individual AI vendors cannot replicate without harming their own competitive positions. - The technical decision to embed graph_data in URLs is strategically critical - it transforms simple document sharing into a viral growth mechanism that provides more value to recipients than native platform alternatives. - Success depends on seamless integration into existing AI workflows rather than requiring users to adopt new tools, with each platform needing tailored integration patterns while maintaining consistent URL behavior. ## Open questions / gaps - Performance metrics and user adoption data to validate the theoretical competitive advantages described in the strategy document. - Technical specifications for the graph_data format and versioning strategy to ensure long-term URL stability. - Detailed analysis of potential competitive responses from major AI vendors and defensive strategies. - User experience research on how different integration patterns affect adoption and retention across platforms. ## Notable connections - **doc:rBRZ_nni** ↔ **doc:kxn9T-Vv** — The strategic moat theory depends entirely on the technical implementation of graph_data embedding - without rich URL payloads, there's no viral loop advantage. - **doc:kxn9T-Vv** ↔ **doc:TEuiwop8** — The bundle URL decision enables practical integrations like Cursor - the technical choice directly impacts real-world usability patterns. - **doc:rBRZ_nni** ↔ **doc:TEuiwop8** — Cross-AI portability strategy manifests in practice through integrations like Cursor that demonstrate consistent URL behavior across different AI platforms. ## Concepts (this bundle) - **Cross-AI Portability** - **Viral Loop** - **Structural Moat** - **URL Contract** - **Analysis Staleness** - **Privacy Gating** - **AI Memory** - **Vendor Neutrality** ## Concept relations - **Cross-AI Portability** ↔ **Structural Moat** — creates - **Structural Moat** ↔ **Vendor Neutrality** — requires - **Viral Loop** ↔ **Cross-AI Portability** — depends on - **URL Contract** ↔ **Cross-AI Portability** — enables ## Documents ### 1. [The structural moat: cross-AI portability](https://memory.wiki/rBRZ_nni) > Why cross-AI portability is the moat. Read first. A single AI vendor can build deeper integration against its own model than mdfy ever could. None of them can deliver a URL that works across their competitors. The portability is the product. *sections:* One line summary: > A single AI vendor can build deeper integration against its own model than mdfy ever could. None of them can deliver a URL that works across their competitors | Why this matters: Notion, Mem.ai, Roam, Obsidian — each is a destination. The user is asked to live inside the tool. mdfy is the opposite shape: the user lives wherever they alre | What gets ported: The doc body (clean markdown); The graph analysis (themes, insights, concept relations) attached to bundles; The concept index attached to hubs; Privacy gating (Public / Restricted / Private) — the URL behaves the same way the rendered viewer does | … ### 2. [Decision: ship graph_data inside the bundle URL](https://memory.wiki/kxn9T-Vv) > The ADR that operationalises the strategy. **Status**: shipped 2026-05-12 (commit `ba7344c4`) *sections:* Context: A bundle URL returns a markdown digest. Before this change the digest contained only the doc list + annotations. The canvas analysis (themes, insights, concept | Decision: Embed the graphdata JSON as markdown sections inside the /raw/bundle/ response by default. Add ?graph=0 opt-out for callers that want the legacy doc-list-on | Why: The viral loop runs through URLs. If pasting a URL gives the receiving AI more than the sending AI could give without mdfy, the URL is genuinely portable. Witho | … ### 3. [Wiring mdfy into Cursor](https://memory.wiki/TEuiwop8) > Concrete tool wiring — the URL shape this ADR produces flowing into Cursor. `.cursor/rules/mdfy.mdc`: *sections:* How Cursor uses this: On every new chat or composer session, Cursor evaluates files matched by the rule and includes them as context. When mdfy URL is in the rule body, Cursor fetche | What we observed in practice: First few messages: Cursor doesn't always fetch — it relies on its own RAG over the local repo. Fine.; Once the user types something topic-specific ("how did we decide the pricing tier?"), Cursor fetches the bundle URL and the answer comes back cited to the right; Cursor's [doc:N] citations don't propagate the way Claude Chat's do — Cursor inlines snippets but doesn't link back. Acceptable for a code-editor surface. | … _Digest view — follow any link above to fetch that doc's full markdown. Add `?full=1` to this URL for the concatenated payload._