First hour with LQ.AI

1.3 Transparency as a Founding Principle

LQ.AI's commercial competitors treat their prompt engineering as proprietary moat. The skills, playbooks, citation logic, and verification heuristics that shape what the user sees are hidden inside closed-source applications, presented as "AI" but functionally indistinguishable from "a system prompt the vendor refuses to show you." This is smoke and mirrors. Much of the time, the emperor has no clothes — what looks like advanced legal AI is a moderately well-tuned prompt that the vendor charges hundreds of dollars per seat per month to keep secret.

LQ.AI inverts this. Every artifact that shapes the user's experience is visible work product. The skills are open source. The playbooks are open source. The citation engine's verification logic is open source. The Enhance Prompt rewriter is open source. The autonomous-agent instructions are open source. The Organization Profile (§3.12) — the org-wide voice, templates, and "what good looks like" reference that shapes every output — is open source. When a user clicks "view this skill" on any active skill, they see the actual SKILL.md and supporting files, formatted for human reading, with provenance and the ability to fork. There is no hidden layer between the user's prompt and the model's output that the user cannot inspect.

This commitment shapes three concrete product decisions:

1. No proprietary "secret sauce" in the open-source release. Optimizations that depend on undisclosed prompt engineering, undisclosed routing rules, or undisclosed verification heuristics are not part of LQ.AI. If we use a clever technique, the technique is in the repo, documented, and contributable.
2. Skill inspectability is a first-class application feature (§3.4), not a developer-debug affordance. Every active skill is one click away from being readable. Users learn the patterns, build trust through verification, and disagree-fork-replace when the skill is wrong.
3. The skills are the product. The value of LQ.AI comes from the curation and authoring of skills — which the community can read, contribute to, and improve — not from hiding them behind a paywall. Skills written for LQ.AI work in any agentskills.io-compatible runtime; users are not locked in.

The position implied by all of this is uncomfortable for the rest of the legal-AI category, and that is intentional. Customers who have been paying significant per-seat fees for software whose only real innovation is a hidden system prompt are entitled to see what they have actually been buying. When the curtain is pulled back, some of those products will hold up. Many will not. LQ.AI's bet is that an open, transparent product built on community-curated skills is better than a closed, opaque product built on the assumption that the user cannot see what is happening — and that the resulting trust is worth more than the marketing.

Practical implication for contributors and operators: treat skills as the canonical artifact. When something in the system produces a wrong answer, the answer to "why" is almost always in a SKILL.md somewhere. When the right answer is something we want the system to do consistently, the way to get there is to write or improve a skill. Skills are not configuration; they are the substance of the product.