feat(ai): Phase 2 — chat assistant with tool-calling

Second phase of the AI feature. Users can now open a slide-in chat drawer from the SiteHeader and ask natural-language questions about their company's invoices, customers, payments, and expenses. The LLM invokes pre-defined read-only tool functions (scoped to the current company at execute time) to fetch data and synthesize answers.

**Database** — new ai_conversations and ai_messages tables. Messages are stored in OpenAI's chat format so AiAssistantService serializes a conversation into an API request with zero translation. Columns: role, content, tool_call_id, tool_calls JSON, model, tokens_in, tokens_out. Conversations are scoped (company_id, user_id) — one user's chats are invisible to everyone else, even inside the same company. Foreign-key cascade deletes.

**Tool infrastructure** — AiTool abstract base + AiToolRegistry singleton (registered in a new AiServiceProvider). The base class enforces the security rule: every tool's execute() receives companyId and userId as injected parameters; tools' JSON schemas NEVER include a company_id field. An LLM physically cannot pass a company_id and escape tenancy. Modules can register their own tools by resolving the registry from their own ServiceProvider::boot().

**Nine built-in tools**: search_invoices, get_invoice, search_customers, get_customer, list_recent_payments, list_overdue_invoices, get_company_stats (aggregates for named periods), search_items, list_expense_categories. All read-only; no mutations. Each returns JSON-encodable data the LLM can parse.

**AiAssistantService orchestration loop** — the heart of Phase 2. Flow: persist user message → build payload from system prompt + recent history (40-message window) + new user message → call driver.chatCompletion with tools → if tool_calls, execute each one via the registry (with injected scope), persist tool result, loop → if plain text, persist and return. Hard cap at 5 iterations to prevent runaway LLMs. System prompt pins the assistant to this company's data and forbids mutation.

**Controllers + policy + rate limit** — POST /api/v1/ai/chat runs the orchestration loop. GET/PATCH/DELETE /api/v1/ai/conversations for CRUD. AiConversationPolicy enforces user_id+company_id match on every action. A new 'ai' RateLimiter in RouteServiceProvider throttles to 30 req/min per (user, company). New 'use ai' Gate defined in AppServiceProvider returns true for any authenticated user — the per-company kill-switch still goes through AiConfigurationService::resolveForCompany.

**Frontend** — new features/company/ai/ folder with a Pinia store (ai-chat.store.ts) holding drawer state, current conversation, messages, and loading flags. AiChatDrawer.vue is a slide-in panel teleported to <body>, mounted globally in CompanyLayout.vue when bootstrap reports ai.enabled && ai.chat_enabled. Sub-components: AiChatMessage (user bubbles vs assistant bubbles), AiChatMessageInput (Enter submits, Shift+Enter newline), AiChatConversationList (sidebar with 'new chat' button, rename, delete). A SparklesIcon button in SiteHeader toggles the drawer.

**Driver test double** — tests use a ScriptedAiDriver registered via AiDriverFactory::register('scripted', ...) that returns pre-queued AiChatResponse objects. Feature tests cover: happy path (new conversation + message persistence), tool-call loop (multi-round-trip with search_invoices), runaway-loop cap, driver-throws path, ai_enabled=NO rejection, chat role disabled rejection, per-user conversation visibility, cross-user policy enforcement, cascade delete.

388 tests pass (was 372, +16 new). Pint clean. npm run build clean. Phase 3 (WYSIWYG text generation popup) is the remaining follow-up.
This commit is contained in:
Darko Gjorgjijoski
2026-04-12 08:00:00 +02:00
parent c7fab5d52f
commit e861fc1fc1
39 changed files with 2776 additions and 1 deletions

View File

@@ -0,0 +1,79 @@
<script setup lang="ts">
import { useAiChatStore } from '../stores/ai-chat.store'
import type { AiConversationSummary } from '@/scripts/types/ai-config'
const store = useAiChatStore()
async function select(convo: AiConversationSummary): Promise<void> {
await store.loadConversation(convo.id)
}
async function remove(convo: AiConversationSummary, event: MouseEvent): Promise<void> {
event.stopPropagation()
if (!window.confirm('Delete this conversation?')) return
await store.deleteConversation(convo.id)
}
</script>
<template>
<div class="flex flex-col h-full">
<div class="p-3 border-b border-line-default">
<button
type="button"
class="w-full text-left text-sm font-medium rounded px-3 py-2 bg-btn-primary text-white hover:bg-btn-primary-hover"
@click="store.newConversation()"
>
+ {{ $t('ai.chat.new_conversation') }}
</button>
</div>
<div class="flex-1 overflow-y-auto">
<div
v-if="store.isLoadingConversations && store.conversations.length === 0"
class="p-3 text-xs text-muted"
>
{{ $t('general.loading') }}...
</div>
<div
v-else-if="store.conversations.length === 0"
class="p-3 text-xs text-muted"
>
{{ $t('ai.chat.no_conversations') }}
</div>
<ul v-else class="space-y-1 p-2">
<li
v-for="convo in store.conversations"
:key="convo.id"
>
<button
type="button"
class="
w-full text-left flex items-center justify-between
px-3 py-2 rounded text-sm group
hover:bg-hover
"
:class="{
'bg-hover-strong font-semibold': store.currentConversationId === convo.id,
}"
@click="select(convo)"
>
<span class="truncate text-body">
{{ convo.title ?? $t('ai.chat.untitled') }}
</span>
<span
class="
ml-2 text-xs text-muted opacity-0 group-hover:opacity-100
hover:text-alert-error-text
"
@click="remove(convo, $event)"
>
{{ $t('general.delete') }}
</span>
</button>
</li>
</ul>
</div>
</div>
</template>

View File

@@ -0,0 +1,140 @@
<script setup lang="ts">
import { nextTick, ref, watch } from 'vue'
import { useAiChatStore } from '../stores/ai-chat.store'
import AiChatMessage from './AiChatMessage.vue'
import AiChatMessageInput from './AiChatMessageInput.vue'
import AiChatConversationList from './AiChatConversationList.vue'
const store = useAiChatStore()
const messagesEl = ref<HTMLDivElement | null>(null)
// Auto-scroll to the bottom whenever the message list grows.
watch(
() => store.messages.length,
async () => {
await nextTick()
if (messagesEl.value) {
messagesEl.value.scrollTop = messagesEl.value.scrollHeight
}
},
)
async function onSend(message: string): Promise<void> {
await store.sendMessage(message)
await nextTick()
if (messagesEl.value) {
messagesEl.value.scrollTop = messagesEl.value.scrollHeight
}
}
</script>
<template>
<!-- Backdrop -->
<Teleport to="body">
<transition name="ai-drawer-fade">
<div
v-if="store.isOpen"
class="fixed inset-0 bg-black/20 z-40"
@click="store.close()"
/>
</transition>
<!-- Drawer panel -->
<transition name="ai-drawer-slide">
<aside
v-if="store.isOpen"
class="
fixed top-0 right-0 bottom-0 z-50
w-full sm:w-[480px] lg:w-[640px]
bg-surface shadow-2xl
flex
"
>
<!-- Conversation list sidebar -->
<div class="hidden sm:block w-48 border-r border-line-default bg-surface-secondary">
<AiChatConversationList />
</div>
<!-- Messages + input -->
<div class="flex-1 flex flex-col">
<div class="flex items-center justify-between p-3 border-b border-line-default">
<div class="flex items-center gap-2">
<BaseIcon name="SparklesIcon" class="w-5 h-5 text-primary-500" />
<h2 class="text-sm font-semibold text-heading">
{{ $t('ai.chat.title') }}
</h2>
</div>
<button
type="button"
class="text-muted hover:text-heading"
@click="store.close()"
>
<BaseIcon name="XMarkIcon" class="w-5 h-5" />
</button>
</div>
<div
ref="messagesEl"
class="flex-1 overflow-y-auto p-4 space-y-3"
>
<div
v-if="store.messages.length === 0"
class="text-center text-sm text-muted mt-12"
>
<BaseIcon name="SparklesIcon" class="w-10 h-10 mx-auto mb-2 text-subtle" />
<p>{{ $t('ai.chat.empty_state') }}</p>
</div>
<AiChatMessage
v-for="msg in store.messages"
:key="msg.id"
:message="msg"
/>
<div
v-if="store.isSending"
class="flex justify-start"
>
<div class="bg-surface-tertiary rounded-lg px-4 py-2 text-sm text-muted italic">
{{ $t('ai.chat.thinking') }}
</div>
</div>
<div
v-if="store.lastError"
class="p-3 text-xs text-alert-error-text bg-alert-error-bg rounded"
>
{{ store.lastError }}
</div>
</div>
<AiChatMessageInput
:is-sending="store.isSending"
@send="onSend"
/>
</div>
</aside>
</transition>
</Teleport>
</template>
<style scoped>
.ai-drawer-fade-enter-active,
.ai-drawer-fade-leave-active {
transition: opacity 0.2s ease;
}
.ai-drawer-fade-enter-from,
.ai-drawer-fade-leave-to {
opacity: 0;
}
.ai-drawer-slide-enter-active,
.ai-drawer-slide-leave-active {
transition: transform 0.25s ease;
}
.ai-drawer-slide-enter-from,
.ai-drawer-slide-leave-to {
transform: translateX(100%);
}
</style>

View File

@@ -0,0 +1,31 @@
<script setup lang="ts">
import { computed } from 'vue'
import type { AiChatMessage } from '@/scripts/types/ai-config'
const props = defineProps<{
message: AiChatMessage
}>()
const isUser = computed(() => props.message.role === 'user')
</script>
<template>
<div
class="flex"
:class="isUser ? 'justify-end' : 'justify-start'"
>
<div
class="max-w-[85%] rounded-lg px-4 py-2 text-sm"
:class="
isUser
? 'bg-primary-500 text-white'
: 'bg-surface-tertiary text-body'
"
>
<!-- Plain text rendering for v1 markdown support can ship later. -->
<p class="whitespace-pre-wrap break-words">
{{ message.content ?? '' }}
</p>
</div>
</div>
</template>

View File

@@ -0,0 +1,62 @@
<script setup lang="ts">
import { ref } from 'vue'
const props = defineProps<{
isSending?: boolean
}>()
const emit = defineEmits<{
send: [message: string]
}>()
const text = ref<string>('')
function submit(): void {
const trimmed = text.value.trim()
if (!trimmed || props.isSending) return
emit('send', trimmed)
text.value = ''
}
/**
* Shift+Enter → newline, Enter alone → submit (standard chat UX).
*/
function onKeydown(e: KeyboardEvent): void {
if (e.key === 'Enter' && !e.shiftKey) {
e.preventDefault()
submit()
}
}
</script>
<template>
<form
class="border-t border-line-default p-3 flex items-end gap-2"
@submit.prevent="submit"
>
<textarea
v-model="text"
rows="2"
class="
flex-1 resize-none rounded-md border border-line-default
bg-surface text-body text-sm px-3 py-2
focus:outline-none focus:ring-1 focus:ring-primary-500
"
:placeholder="$t('ai.chat.input_placeholder')"
:disabled="isSending"
@keydown="onKeydown"
/>
<button
type="submit"
class="
rounded-md px-3 py-2 text-sm font-medium
bg-btn-primary text-white hover:bg-btn-primary-hover
disabled:opacity-50 disabled:cursor-not-allowed
"
:disabled="!text.trim() || isSending"
>
{{ isSending ? $t('ai.chat.sending') : $t('ai.chat.send') }}
</button>
</form>
</template>