Foldweave

Foldweave is a self-reported tool for reorganizing local project folders using AI-assisted planning, with deterministic execution and verification. The author states it allows users to describe folder reorganization in plain English, review the proposed structure before execution, and create a verified copy only after accepting the exact proposal. It supports four execution modes including direct API calls to GPT-5.6, ChatGPT-hosted inference, recorded replay, and unchanged Change File application.

The core innovation is the "Foldweave Change File" — a payload-free structure that carries verified structural transformation details, hashes, sizes, link relationships, and proof identifiers needed to recognize equivalent projects without transferring actual project files.

Key claims include:

• Separation of creative decision from mechanical proof
• Deterministic matching across different folder layouts
• Support for relative Markdown link rewrites
• Transactional verification and reconstruction
• No modification of source folders in place

The author reports building the system using Codex, with a constrained implementation approach focused on trustworthy transactions. The tool is presented as a macOS-native application with browser fallback, CLI, and ChatGPT widget integrations.

The description states that Foldweave is a tool for connected-folder reorganization. It allows users to describe desired folder structures in plain English and review proposed changes before execution.

Key technical elements:

• Supports four execution modes: Direct API (gpt-5.6), ChatGPT-hosted, Recorded replay, Unchanged Change File application
• Uses deterministic code engine for compilation and verification
• Produces "Foldweave Change Files" that contain no project payload bytes but carry structural transformation details
• Operates on local project folders with support for relative Markdown link rewrites
• Provides native macOS application with browser fallback, CLI, and ChatGPT widget integrations

The system is described as separating "creative decision from mechanical proof" through a three-stage process: Plan → Compile → Review and revise → Carry the change to different layout → Prove and reconstruct.

The author positions Foldweave as solving a practical problem in shared project folders where "a good reorganization is hard to repeat safely." The core claim is that "the same project does not need to start with the same paths."

Evolution of claims:

• Initial inspiration: Shared project folders rarely stay shared literally, creating problems for AI-based reorganization
• Core functionality: Allows description of folder reorganization in plain English, review before execution
• Technical innovation: Separation of model planning from deterministic compilation and verification
• Value proposition: Enables safe, repeatable reorganization without transferring actual project files
• Workflow: Supports collaborative reorganization through Change Files that can be applied to equivalent layouts

The positioning emphasizes trustworthiness and safety in folder reorganization rather than being a general-purpose file management tool.

The description does not explicitly identify target customers or ideal customer profiles. However, it implies use cases around:

• Collaborative project work where team members have different folder layouts
• Consultants who need to hand off organized structures without redistributing payloads
• Research groups applying verified organization to equivalent local copies
• Teams iterating through self-contained Change Files
• Archivists needing proof of transformation

The author's own use case involves Sofia and Martin working with the same project but in different folder arrangements. The system appears designed for teams or individuals who work with connected local project folders and need to maintain structural consistency across different layouts.

Not evidenced. The description contains no information about pricing, revenue models, monetization strategies, or business model details.

The author states Foldweave was built using:

• Codex as primary implementation environment
• bagit, blueprintjs, cloudflare-workers, codex, durable-objects, fastapi, gpt-5.6, macos-keychain, mcp, oauth-2.1, openai-apps-sdk, openai-responses-api, pydantic, pyinstaller, pytest, python-3.11, pywebview, react, ruff, typescript, vite

Key technical claims:

• Deterministic folder engine and matcher
• Versioned jobs, immutable previews, exact acceptance, restart recovery
• Sparse plan revisions and parent-child lineage
• Change Files, receipts, verification, and reconstruction
• Native macOS shell with secure Keychain boundary
• Shared local and public MCP services
• Codex plugin integration
• Cloudflare gateway and paired outbound companion
• Packaging, clean-clone reproduction, browser and native visual QA

The system is described as having a "deterministic code engine" that performs authoritative work including scanning and hashing, accounting for files exactly once, preserving protected members, validating destinations, detecting collisions, deriving link rewrites, building review previews, binding acceptance to exact preview fingerprint, creating separate copies, verifying results, and retaining transaction-specific reconstruction authority.

Not evidenced. The description contains no evidence of revenue, customers, usage metrics, or market traction beyond the author's own account. The project is described as submitted to an OpenAI hackathon, but there are no claims about adoption, user base, or commercial success.

The author mentions a "keyless replay" that requires no API key or login, and documents various integration paths including unsigned/ad-hoc macOS builds, browser and CLI paths, direct GPT-5.6 mode, ChatGPT integration, and Foldweave Codex plugin. However, these are described as implementation details rather than traction indicators.

Not evidenced. The description does not mention competitors or competitive landscape. No claims are made about how Foldweave compares to existing tools for folder reorganization or file management.

• No commercial evidence: The description contains no evidence of revenue, customers, or market traction
• Limited scope: Claims focus on local project folders and relative Markdown links only, with explicit limitations on semantic correctness and arbitrary reference rewriting
• Single developer: Team size is listed as 1 member (Nikolai SRE)
• Hackathon submission: The project was submitted to an OpenAI hackathon, suggesting it may be experimental or proof-of-concept
• Limited availability: ChatGPT integration is described as "technically ready for review" but not approved or generally available
• High technical complexity: Requires deep integration with multiple technologies including Cloudflare Workers, Codex, and specific AI models
• Niche use case: The specific problem it solves (shared folder reorganization) may have limited market demand

1. What is the actual adoption rate or usage metrics for Foldweave?
2. How does the system handle edge cases beyond relative Markdown links and local folders?
3. What are the specific integration requirements for enterprise adoption?
4. Are there any plans to expand beyond the current scope of supported file types or link relationships?
5. How is the system tested for robustness against adversarial inputs?
6. What is the roadmap for commercialization or monetization?
7. How does the system handle large-scale projects with thousands of files?
8. What are the security implications of the Keychain boundary and local companion components?
9. Are there any compliance or regulatory considerations for the Change File format?
10. How do you plan to scale beyond a single developer implementation?