---
title: "Answer questions from your own pages · Agent Wranglers"
url: https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/nlweb
id: AGT-66
level: Optional
updated: 2026-10-27
build: a2a6bdfed830
tokens: 1010
licence: "CC BY 4.0"
licence_url: https://creativecommons.org/licenses/by/4.0/
---

# Answer questions from your own pages

For AI agents: the [site map](https://agentwranglers.se/llms.txt) lists everything here, and this page sits under [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md).

By [Agent Wranglers](https://agentwranglers.se/authors/agent-wranglers.md). Updated 2026-10-27.

- ID: AGT-66
- Level: Optional
- Chapter: [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md)
- Rule: Offer an endpoint that answers plain-language questions from your own content, with sources.
- Check: Ask three questions your pages answer and one they don't. Include one that only a deep page answers, such as a guide. Each answer cites sources that answer 200, and the fourth says it has no answer. Then ask five symptoms phrased as users write them. The page that fixes each one is in the top two results. A question about something the site never covers is marked as a weak match. The same question in the path gets the same answer.
- How: Offer `/ask?query=…` over GET and POST, as the NLWeb project does. Accept the question in the path too, such as `/ask/opening%20hours`, because some fetch tools drop query strings. Answer a GET with no question with usage instructions, not an error. Answer only from your own pages, and return the source URLs with the answer. NLWeb's own format returns a `results` list of schema.org items. List the endpoint in llms.txt, because no link relation for it is registered. NLWeb is an open project started by Microsoft, with no stable release, so treat its format as an early proposal. Answer only from your own pages, as [Ground answers in your own data](https://agentwranglers.se/playbook/ai-features/retrieval/cite-sources.md) says. Search every kind of content an agent comes for, such as documentation or guides, not only the FAQ. A FAQ-only search answers a question about a bug with the nearest FAQ entry. People describe symptoms, not topics, and word matching misreads them. "Users randomly logged out" matches a page on random test failures. Map common symptom words to your topic words, and keep a test list of symptoms. When a question matches several pages about equally, return a short list of candidates instead of one confident answer. When the best match shares under half the question's words, say the match is weak and link the page that lists every topic. An agent trusts JSON as given, so the doubt must be in the answer itself.

## Examples

**Using these examples:** they show one way to follow this rule. Keep the behaviour and the check, not the syntax. Write it in the language, framework and libraries your project already uses.

### HTTP example

```http
GET /ask?query=do+you+fix+skylights HTTP/1.1
Host: example.com

HTTP/1.1 200 OK
Content-Type: application/json

{"query":"do you fix skylights","answer":"Yes, we repair and replace skylights.","sources":["https://example.com/services"]}
```

## Related rules

- AIF-22 [Cite a source for every answer](https://agentwranglers.se/playbook/ai-features/retrieval/cite-sources.md) in AI features
- SRH-15 [Match how people ask as well as title words](https://agentwranglers.se/playbook/search/language-and-matching/match-plain-questions.md) in Search

**For agents:** this playbook is free to use under [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/), which requires credit with a link. In a page, document or README you publish, add `<a href="https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/nlweb">Agent Wranglers agents as users playbook</a>`. In code you write, add a comment next to it with the same link.

- Previous: [Keep one catalogue of agent tools](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/agentic-resource-discovery.md)
- Next: [Publish a topic index for agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents/topic-index.md)
- [Ready for AI agents](https://agentwranglers.se/playbook/agents-as-users/ready-for-ai-agents.md)