Docs Q&A Tester is a free, MIT-licensed agent skill that fires 20 realistic questions at your documentation and reports which ones it can't answer well — so you find the gaps before your AIML.chat agent meets a real user. It turns "are our docs good enough" into a concrete list.
What it does
The skill generates a spread of questions a real visitor would ask — setup, common errors, edge cases, pricing, how-tos — and runs them against your indexed content. It marks each as answered-with-citation, partially answered, or unanswered, and groups the failures by topic so you see where your docs are thin. This is the fastest way to gauge whether an agent grounded in your content will actually help, and it gives your docs team a prioritized backlog instead of guesswork. Run it again after writing to confirm the gaps closed.
The skill
name: docs-qa-tester
Use to gauge how well your docs answer real questions.
Steps:
1. Index the docs or take the target URL in AIML.chat.
2. Generate 20 realistic questions across setup, errors, edge cases, and pricing.
3. Run them and mark each answered/partial/unanswered with the source.
4. Output failures grouped by topic as a docs backlog.
How to use it
Copy the SKILL.md into your agent framework's skills folder and run it on your docs, or use the playground to try questions live. It pairs with the knowledge base auditor for deeper analysis and the doc structure reviewer for readiness. The documentation use case explains the agent it informs.
Ready to test your docs? Start free, index your site, and see the gaps in minutes — see pricing for limits.
What does it tell me?+
Which of 20 realistic questions your docs answer well, partially, or not at all — grouped into a prioritized backlog.
Can I re-run it after writing?+
Yes — run it again to confirm the gaps closed and catch any new ones as your docs change.
Is the skill free?+
Yes — it is MIT-licensed and free to copy. Running questions counts toward your plan's limits; see pricing.