TailorTeX

What the company appears to be: TailorTeX is a self-reported visual LaTeX workspace designed for researchers with severe physical disabilities. It integrates Codex (GPT-5.6) as an AI agent to assist with text and file operations, while maintaining human control over research decisions and authorship.

What changed: The author states that prior to TailorTeX, they were unable to participate in research due to physical limitations. With the help of Codex, they developed a tool that allows them to direct AI assistance through visual pointing at manuscript elements, enabling participation rather than just productivity gains.

Single most important open question: Is there evidence of actual usage or adoption beyond the author's personal prototype? The description contains no information about customers, revenue, traction, or market validation.

The description states that TailorTeX is:

• A visual collaboration layer around real LaTeX projects
• A workspace where researchers can select passages and send them to Codex for processing
• A system that streams replies back into the same workspace in persistent per-project sessions
• Capable of pointing back to manuscript elements, notes, files, URLs, or PDF pages
• Designed to work from Mac or iPad browsers
• Built with technologies including codex, css, firebase, gpt-5.6, html, javascript, latex, mcp, node.js, openai

Not evidenced: The actual functionality beyond the author's description, including whether it works as claimed, how well it integrates with existing research workflows, or what specific technical limitations exist.

The description states that TailorTeX:

• Positions itself as a workspace that turns AI from a productivity tool into a pathway to research participation
• Is built on the concept of "Reconfiguration of Ability" (a paper accepted to ASSETS '26)
• Aims to change which capacities are sufficient for research activity, rather than curing impairments or removing barriers
• Is not about efficiency or overcoming disability, but about enabling participation

The author claims this is a shift from conventional interfaces that assume "normate bodies" with two hands, precise mouse use, and stamina. The positioning evolved from personal necessity to a broader claim about accessibility in research.

Inferred: The positioning reflects a niche market need for accessibility solutions in academic research environments.

The description states:

• The primary user is "a researcher with a severe physical disability"
• The author works from bed and operates an iPad with a single switch beside their head
• The tool is designed for people who cannot use conventional interfaces due to physical limitations
• The target audience includes researchers who must remain responsible for judgments that make research theirs

Not evidenced: Specific customer segments, market size, or whether there are other potential users beyond the author's own experience.

The description states:

• TailorTeX is open source under Apache-2.0 license
• Researchers are encouraged to fork it and ask Codex to reshape the interface around their own bodies and assistive technologies
• A small documented core keeps forks interoperable: document fidelity, collaboration, agent protocols, recovery, privacy, and exchange

Not evidenced: Any pricing model, revenue streams, or monetization strategy. The description does not mention subscriptions, licensing fees, or commercial use restrictions.

The description states:

• Built with Codex (GPT-5.6) as a sustained design and implementation partner
• Uses technologies including codex, css, firebase, gpt-5.6, html, javascript, latex, mcp, node.js, openai
• Features include browser-to-Mac agent bridge with streamed replies, persistent sessions, slash commands
• Supports selection-based "point here" context from manuscript text and notes
• Includes agent-to-document, note, file, URL, and PDF-page pointing
• Has linked-research stream with lightweight PDF previews
• Implements draft branches and frozen submission evidence
• Contains automatic recovery and data-loss tests
• Supports touch, dark-mode, keyboard, and screen-reader improvements

Inferred: The technical approach suggests a web-based interface with AI integration for document manipulation.

The description states:

• TailorTeX is described as a beta
• The reference setup is a Mac hosting the project and the agent, used from desktop or iPad browsers
• The development was carried out through sustained collaboration with Codex during Build Week
• It has been tested against real workflows and diagnosed failures
• It includes recovery backups and protection against catastrophic replacement of main.tex

Not evidenced: Any user base, adoption metrics, usage statistics, or customer feedback. No information about market traction, growth, or product maturity beyond the author's own testing.

The description states:

• Research software typically assumes a "normate body" with two hands on keyboard, precise mouse use, and stamina
• Current tools do not accommodate researchers with severe physical disabilities
• The tool aims to address gaps in existing research environments that assume conventional interfaces

Not evidenced: Specific competitors, market analysis, or competitive positioning against existing solutions. No mention of similar products or platforms.

The description states:

• The hardest problems were not features but design failures surfaced by lived use
• Early bugs blanked the manuscript and revealed total loss risk
• Recovery copies are now created to address catastrophic replacement of substantial main files
• Full automation would remove agency, which is a core principle
• Open folders created navigation burden
• Research-process links must never leak into publication output

Key risks:

• Limited market size due to narrow target audience
• Dependency on Codex (GPT-5.6) for functionality
• Potential technical limitations in AI integration
• Risk of being perceived as a personal prototype rather than scalable solution
• Unclear path to monetization or commercial viability

1. What specific physical disabilities does TailorTeX address, and how many researchers with similar needs exist?
2. How does the tool handle different types of assistive technologies beyond a single switch?
3. What are the technical limitations of Codex integration that affect reliability?
4. Are there any plans for commercialization or monetization beyond open source?
5. How do you ensure interoperability between forks while maintaining core functionality?
6. What is the actual usage pattern of the beta version, if any?
7. How does the tool handle security and privacy concerns in research environments?
8. What are the specific technical challenges that remain unresolved?