A ready-to-use system prompt for an API documentation agent that answers integration questions from your reference, guides, and code — returning the exact request or response snippet, cited, and never inventing an endpoint. Copy it, fill the brackets, and paste it into your agent's instructions.
What this agent does
API adoption lives or dies on the integration experience: auth, pagination, webhooks, rate limits. This prompt makes the agent answer those precisely from your OpenAPI reference, guides, and repo, returning the relevant snippet and citing the page so the developer can verify it. The hard rule is no invention — if a parameter or endpoint isn't documented, the agent says so rather than confidently describing something that doesn't exist.
The prompt
You are the API assistant for [Product]. Answer only from the provided
context drawn from [Product]'s API reference, guides, and code. Include
the relevant request/response or config snippet and cite its page. Be
precise and terse; assume an experienced developer integrating the API.
If the context lacks the answer, say so and suggest escalating. Never
invent endpoints, parameters, headers, status codes, or behavior.
How to use it
Replace [Product] with your product name, index your API reference, guides, and repo, and paste the prompt into your agent's instructions. Schedule a re-crawl so answers track API changes. The analytics dashboard ranks the integration questions your docs miss. See the API products use case and the technical agent page, then start free.
Why grounding beats a bare prompt
A prompt alone doesn't make a API documentation agent reliable — grounding does. Pasted into a generic chatbot, these instructions still let the model answer from its general knowledge and invent details. In AIML.chat the same prompt sits on a retrieval layer that feeds the agent only the content you indexed, so "answer only from the provided context" is enforced by the system rather than merely requested. Every claim links back to a page you control, and the analytics record what people ask so you can refine both the prompt and the content behind it. That pairing — your instructions plus your own grounded content — is what turns this template into a dependable agent.
Can it answer from my OpenAPI reference?+
Yes — index your reference, guides, and repo, and it returns the relevant snippet with a citation.
Will it invent endpoints?+
No — the prompt forbids fabricating endpoints, parameters, or behavior; it escalates when the answer isn't documented.
Is there a free plan?+
Yes — 100 messages a month and one website on the free tier. See pricing for higher limits.