feat: suggest dataservice description

[bolinocroustibat]

Add AI-powered description suggestion for dataservices (external APIs) via the Albert API.

When editing a dataservice, a "Suggérer une description" button generates a French description. The button is enabled only when:

• title is filled
• at least one of these is filled: technical documentation URL or machine documentation URL (OpenAPI/Swagger)

Changes:

• DescribeDataservice.vue: suggestion button with loading/disabled states and tooltip; enabled when title + at least one doc URL are present
• generate-dataservice-description.post.ts (new): Nitro endpoint that fetches documentation from the given URLs, inlines it into the prompt, and calls Albert API. Requires title and at least one of technicalDocumentationUrl or machineDocumentationUrl
• fetch-documentation.ts (new): utility to fetch doc content from URLs (HTML stripped, JSON/YAML as-is; 15s timeout, 120k char cap)

---

EDIT (2026-02-17):

• generate-dataservice-description.post.ts uses shared callAlbertAPI (albert-helpers), same pattern as other Albert endpoints.
• Doc content is now fetched from the URLs and inlined into the prompt (new fetch-documentation.ts: HTML stripped, JSON/YAML as-is, 15s timeout, 120k char cap). Previously only the URLs were sent.
• Fixed 422 "description too short" being rethrown as 500.
• Removed redundant validateAlbertConfig from helper and all Albert endpoints.

[ThibaudDauce]

Design: The openweight-small model can't browse these URLs — it only sees the URL strings. The prompt asks to "mention key endpoints, data types, and use cases" but the model has no access to the actual documentation content. This will likely produce hallucinated descriptions about specific endpoints.

Possible alternatives:

• Fetch the documentation content server-side and include it in the prompt
• Use createAgentCompletion if the Albert agent API supports web browsing
• Adjust the prompt to only describe what can be reasonably inferred from a title + URL patterns

[bolinocroustibat]

Damned I was tricked by the hallucinations of the models who made me think it was indeed browsing the URLs! Good call.

I would go for option 1, trying to fetch and format the data from those URLs, with some guardrails regarding the maximum size of the prompt and what those models should/could have as a long prompt. I'll suggest a commit soon.

[ThibaudDauce]

Thanks! Really sorry for the delay :-(

[ThibaudDauce]

Could be directly a Promise.all(), no?

Suggested change

	let technicalContent = ''
	let machineContent = ''
	const fetchPromises: Promise<void>[] = []
	if (technicalUrl) {
	fetchPromises.push(
	fetchDocumentationContent(technicalUrl)
	.then((content) => { technicalContent = content })
	.catch((err) => {
	throw createError({
	statusCode: 422,
	statusMessage: err instanceof Error ? err.message : 'Failed to load technical documentation URL.',
	})
	}),
	)
	}
	if (machineUrl) {
	fetchPromises.push(
	fetchDocumentationContent(machineUrl)
	.then((content) => { machineContent = content })
	.catch((err) => {
	throw createError({
	statusCode: 422,
	statusMessage: err instanceof Error ? err.message : 'Failed to load machine documentation URL.',
	})
	}),
	)
	}
	await Promise.all(fetchPromises)
	async function fetchOrFail(url: string, label: string): Promise<string> {
	if (!url) return ''
	try {
	return await fetchDocumentationContent(url)
	}
	catch (err) {
	throw createError({
	statusCode: 422,
	statusMessage: err instanceof Error ? err.message : `Failed to load ${label} documentation URL.`,
	})
	}
	}
	const [technicalContent, machineContent] = await Promise.all([
	fetchOrFail(technicalUrl, 'technical'),
	fetchOrFail(machineUrl, 'machine'),
	])