> For the complete documentation index, see [llms.txt](https://yahya-qanie.gitbook.io/yps-ai-agent/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://yahya-qanie.gitbook.io/yps-ai-agent/limitations-and-responsible-use.md).

# Limitations & responsible use

Know the boundaries before relying on an answer.

### Coverage is bounded

YPS AI Agent draws on a curated collection, not all available YPS knowledge. Some countries, languages, perspectives, and recent developments may be underrepresented. It is not an unrestricted web research service.

### Errors remain possible

The assistant can misread evidence, omit relevant material, produce an unsupported inference, or attach an imperfect citation. Evidence grounding reduces risks but does not make an answer authoritative or error-free.

### Freshness

A source may describe a policy at the time of publication rather than its current implementation. Check recent official information before relying on rapidly changing developments or country-specific requirements.

### Context matters

An approach documented in one country or community may not transfer to another. Adapt ideas with affected communities, local practitioners, and appropriate safeguarding processes.

### Functionality boundaries

* No unrestricted internet search.
* No document uploads.
* No general grammar or text-editing service.
* No user accounts or profiles.
* No professional legal, medical, financial, or emergency advice.

### Use in public or consequential work

Review the original sources before incorporating claims into a policy brief, funding proposal, consultation, or public statement. Be explicit when an idea is your adaptation or an AI-assisted suggestion. Responsibility for decisions remains with the people and organizations using the output.