Friction

The description states that Friction is a tool for analyzing conversations and disagreements to help people understand what they are actually deciding. It allows users to paste a conversation, describe a situation, or enter two different perspectives.

Key features include:

• Finding hidden decisions within situations
• Ranking those decisions by importance
• Explaining what drives each decision
• Comparing options and their benefits/drawbacks
• Identifying whether disagreement is about facts, values, definitions, or missing information
• Suggesting next questions or pieces of evidence
• Saving, sharing, printing, and revisiting decisions

The tool also includes templates, demo scenarios, saved decision processes, a private journal, native sharing, and printable decision briefs.

Evidence Self-reported by the author. No independent verification or demonstration of actual product functionality beyond the description.

The author states that Friction was originally conceived as a "conflict analyser" but evolved into something more focused on helping users understand decisions clearly without pretending an AI can decide who is right.

It positions itself as:

• A decision workspace, not just a text analyzer
• A tool for clarity in disagreement, not resolution
• An assistant that makes uncertainty visible and actionable

The evolution shows a shift from technical complexity to user-focused design, driven by the author's own confusion during testing. The final version focuses on five core steps: Situation, What matters, Options, Analysis, Decision.

Evidence Self-reported evolution narrative. No external validation or data showing how positioning changed in practice.

The description does not name specific target customers or personas. However, the author describes personal experiences with sibling arguments and internship disagreements as inspiration for the product.

It appears aimed at individuals or small groups who:

• Face difficult decisions involving disagreement
• Want to understand what they're actually deciding
• Need clarity in group discussions or conflict resolution

The tool is designed for personal use (e.g., private journal, local storage), though it supports sharing and native integration.

Evidence Self-reported inspiration and intended audience. No evidence of actual customer segments, personas, or market research.

There is no mention of pricing, monetization, or business model in the description. The project is presented as a hackathon submission with no indication of commercial intent or revenue streams.

Evidence Not evidenced.

The frontend was built using React, Vite, and TypeScript. The backend uses Express.js and communicates with GPT-5.6 Luna via REST API. Responses are validated using Zod to prevent malformed outputs from reaching the UI.

Key technical details:

• API key stays on server (no exposure to browser)
• Conversations not stored in database
• Saved data stored locally in user's browser
• Uses Codex for development assistance
• Includes local fallback mode for offline use

Evidence Self-reported technical architecture. No evidence of production deployment, scalability, or performance metrics.

There is no evidence of traction, adoption, or usage beyond the author’s own experience and testing. The project is described as a hackathon submission with no mention of customers, users, or market validation.

Evidence Not evidenced.

The description does not reference any competitors or existing solutions in this space. It does not discuss how Friction compares to other tools for decision-making, conflict resolution, or AI-assisted collaboration.

Evidence Not evidenced.

• No traction or validation: The project is described as a hackathon submission with no evidence of real-world use.
• Unverified claims: All functionality and impact are self-reported without independent verification.
• Single-person team: Only one person built the entire product, raising questions about scalability or long-term maintenance.
• AI dependency: Relies heavily on GPT-5.6 Luna, which may not be available or reliable in production environments.
• Limited scope: The author explicitly states they want to keep it focused, suggesting a lack of ambition for broader market expansion.

Evidence Self-reported claims and assumptions. No external data or third-party validation.

1. What specific real-world situations have you tested Friction on? How did users respond?
2. Have you conducted any usability testing beyond your own experience?
3. Is there a plan to move beyond local storage and into cloud-based data management?
4. How do you intend to monetize or scale this product if it gains traction?
5. What are the limitations of GPT-5.6 Luna in handling complex decision-making scenarios?
6. Are there any plans for team collaboration features, given that the current version is personal-only?