feat: AI-assisted flow builder with 4-stage wizard (#87)
* feat: AI-assisted flow builder with 4-stage wizard Implements the complete AI flow builder feature using a guided 4-stage wizard (Foundation → Scaffold → Branch Detail → Review & Assemble). AI assists at bounded points using Claude Haiku for cost-efficient structured JSON generation (~$0.01-0.03/flow). Backend: new models (ai_conversations, ai_usage), Alembic migration, quota enforcement with billing anchor, Anthropic API integration with prompt caching, tree validation, conversation CRUD with 24h TTL, APScheduler cleanup job, 5 API endpoints, Pydantic schemas. Frontend: TypeScript types, API client, Zustand store for wizard state, 7 components (modal, step indicator, foundation form, branch selector, branch detail view, tree preview, quota display), MyTreesPage integration with "Build with AI" button (hidden when AI not configured). Tests: 14 validator unit tests + 11 endpoint integration tests with mocked Anthropic (zero real API spend). All 25 tests passing. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * docs: dashboard design doc and implementation plan Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: Phase 1 — pinnedFlowsStore, pagination hook, cached quota hook, sidebar refactor - Add pin() to pinnedFlowsApi - Create pinnedFlowsStore (Zustand) — single source of truth for pin state - Add dashboardMyFlowsView preference to userPreferencesStore - Create usePaginationParams hook (URL-synced) - Create useCachedQuota hook (5-min TTL) - Sidebar uses pinnedFlowsStore instead of local state Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: Phase 2 — pin/favorite buttons on all library view components - TreeGridView: star in top-right corner of cards - TreeListView: star at end of each row - TreeTableView: dedicated leftmost Favorite column - All with proper a11y (aria-label), event isolation, loading states Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: Phase 3 — Library page create dropdown + AI Builder + pin wiring - Replace single Create link with dropdown menu (3 flow types + AI Builder) - Wire pinnedFlowsStore to all view components - AI Builder modal integration via useCachedQuota hook Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: Phase 4 — Dashboard refactor with Favorites grid + paginated My Flows - Favorites section: compact grid from pinnedFlowsStore, max 2 rows, expandable - My Flows: author_id filter, URL-synced pagination (10/25/50/All) - View toggle (grid/list/table) with independent preference - Skeleton loaders, empty states with CTAs - Create dropdown with AI Builder option - 500-item ceiling for "Show All" mode Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * feat: Phase 5 — Sidebar pinned section dual collapse + show more/less - Header collapse hides entire section, resets to 5 items on re-expand - List truncation: show first 5, "Show more (N)" expands to all - Clicking a flow auto-collapses back to 5 - Smooth max-height CSS transition (250ms ease-out) Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: stabilize usePaginationParams to prevent infinite re-render loop allowedPageSizes array was recreated every render as a useMemo dep, causing infinite updates. Use useRef to stabilize the reference. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: remove Set-based Zustand selectors causing infinite re-render loop Zustand selectors returning new Set() on every call fail Object.is equality check, triggering continuous re-renders. Replaced with useMemo-derived Sets in consuming components. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: pin route ordering and star icon overlap in grid view Move GET /pinned and PATCH /pinned/reorder before GET /{tree_id} to prevent FastAPI from matching "pinned" as a UUID path parameter (422). Relocate star button from absolute positioning into the header row to avoid overlapping privacy icons and category badges. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: code review fixes — date calc, input validation, rate limits, shared components - Fix monthly_reset_at crash when billing anchor day exceeds next month's length - Add environment_tags sanitization (max 20 tags, 100 chars each) to prevent prompt injection - Add @limiter.limit("10/minute") rate limiting to all AI endpoints - Use getTreeNavigatePath() routing helper instead of hardcoded paths - Extract shared CreateFlowDropdown component from QuickStartPage and TreeLibraryPage - Clear useCachedQuota on logout to prevent stale data across user sessions - Add useRef guard to scaffold useEffect to prevent potential double-fire - Use node.id as React key instead of array index in BranchDetailView - Remove redundant dead logic in ai_tree_validator Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com> * fix: correct Anthropic model ID to full dated version claude-haiku-4-5 is not a valid model alias — Anthropic requires the full dated model ID claude-haiku-4-5-20251001. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: strip markdown code fences from AI JSON responses Haiku sometimes wraps its JSON in ```json ... ``` despite the prompt instructing otherwise. Strip fences before parsing to avoid JSONDecodeError at char 0. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: increase branch_detail max_tokens to 8192 and add response logging Truncated output at 4096 tokens produces invalid JSON mid-generation. Also logs stop_reason and output_tokens per attempt to diagnose failures. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: pass explicit status='draft' when creating AI-generated flow Tree model defaults to 'published' in the DB schema, but passing status=None from the constructor overrides that default, causing a nullable=False violation and a 500 on save. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: auto-advance branch detail and pin navigation bar - Auto-advance to next undetailed branch after generation completes, using a useEffect that watches the count of detailed branches - Cap tree preview at max-h-48 with internal scroll so the nav bar is never pushed off screen - Make nav bar sticky bottom-0 with bg-card so it stays visible regardless of content height Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: increase branch retries to 3 and relax cross-reference validation on final attempt next_node_id mismatches are a common model hallucination that the retry prompt doesn't reliably fix. On the final (3rd) attempt, accept the branch with strict=False so only truly fatal errors (missing fields, dead ends, bad JSON) cause a hard failure. Cross-reference issues are minor and fixable in the tree editor. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: strengthen prompt to prevent next_node_id mismatches, keep strict validation Rather than lowering the validation bar, improve the system prompt: - Rule 6 now explicitly states next_node_id must match a direct child's id - Added rule 10: build tree bottom-up to avoid forward-reference errors - Corrective prompt now calls out the ID mismatch constraint specifically Reverts the strict=False fallback — flows must be correct before saving. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: persist branch viewing index in store to survive phase remounts Local useState resets to 0 every time phase transitions from 'generating' back to 'detailing', causing the view to snap back to branch 1. Move viewingIndex to store's currentBranchIndex (already existed) and advance it in generateBranchDetail after success. Component reads from store so remounts no longer lose position. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: correct publish validation to check title instead of action/solution fields The publish validator was checking for an 'action' field on action nodes and a 'solution' field on solution nodes, but the actual node schema (confirmed from seed data and frontend types) uses 'title'/'description'. This caused all AI-generated trees to fail publish validation. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: correct action node schema and improve AI flow quality - Fix action nodes to use next_node_id (not children) for continuation, matching how TreeNavigationPage.tsx navigates action nodes - Validator now requires next_node_id on all action nodes and flags missing ones as broken dead ends - Update _check_branch_termination: action nodes are not dead ends since they continue via next_node_id (validated separately) - Improve scaffold prompt: branch names must describe observable symptoms users can self-identify, not internal category names - Update branch_detail prompt with clearer action node schema, corrected few-shot example showing proper next_node_id on action nodes - Improve assemble_tree root question to be more user-facing Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * docs: add AI flow builder gotchas to CLAUDE.md (#23-25) - Action nodes use next_node_id (not children) for navigation - Anthropic model IDs require full dated version string - Claude API may wrap JSON in markdown fences Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: resolve CI lint errors and httpx dependency conflict - Fix httpx version conflict: requirements-dev.txt now uses >=0.27.0 to match requirements.txt - Extract CSAT helper functions to csatUtils.ts to fix react-refresh/only-export-components - Remove default export from admin/EmptyState.tsx shim (same rule) - Fix empty catch block in Modal.tsx (no-empty) - Add eslint-disable comments for intentional setState-in-effect patterns in FlowAnalyticsPanel, QuickLaunch, NodeEditorPanel, useCachedQuota, MyAnalyticsPage, TeamAnalyticsPage - Add eslint-disable comments for intentional _children destructure in NodeEditorPanel - Fix _parentId unused var in useTreeLayout.ts - Rewrite usePaginationParams.ts to avoid reading refs during render Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: update tests to match action node schema (next_node_id, not children) - Update _make_valid_tree() in test_ai_tree_validator to use next_node_id on action nodes (solution is a sibling, not a child) - Fix test_dead_end_action_node → test_dead_end_decision_node (action nodes don't have child-based dead ends; dead ends are decision nodes with no children) - Add test_action_missing_next_node_id for the new validation rule - Update BRANCH_DETAIL_JSON in test_ai_endpoints to use next_node_id pattern - Update test_draft_trees.py to use "title" field for action/solution nodes (tree_validation.py was updated this branch to require "title" not "action"/"solution") Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> * fix: update remaining tests and session_to_tree for title field rename - test_tree_validation.py: replace "action"/"solution" content fields with "title" - test_procedural_flows.py: update solution node fixtures to use "title" - test_save_session_as_tree.py: update fixtures and assertions for "title" field - session_to_tree.py: generate "title" instead of "action"/"solution" on converted nodes; fall back to legacy field names when reading from old tree snapshots for compatibility Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com> --------- Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
This commit was merged in pull request #87.
This commit is contained in:
chihlasm
87
backend/app/core/ai_conversation_store.py
Normal file
87
backend/app/core/ai_conversation_store.py
Normal file
@@ -0,0 +1,87 @@
|
||||
"""DB-backed CRUD for AI wizard conversation state.
|
||||
|
||||
Conversations have a 24-hour TTL. Every access validates ownership and expiry.
|
||||
"""
|
||||
import uuid
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from typing import Any, Optional
|
||||
from uuid import UUID
|
||||
|
||||
from fastapi import HTTPException, status
|
||||
from sqlalchemy import select
|
||||
from sqlalchemy.ext.asyncio import AsyncSession
|
||||
|
||||
from app.core.config import settings
|
||||
from app.models.ai_conversation import AIConversation
|
||||
|
||||
|
||||
async def create_conversation(
|
||||
user_id: UUID,
|
||||
account_id: UUID,
|
||||
wizard_state: dict[str, Any],
|
||||
db: AsyncSession,
|
||||
) -> AIConversation:
|
||||
"""Create a new AI wizard conversation."""
|
||||
conversation = AIConversation(
|
||||
user_id=user_id,
|
||||
account_id=account_id,
|
||||
status="foundation",
|
||||
wizard_state=wizard_state,
|
||||
messages=[],
|
||||
expires_at=datetime.now(timezone.utc)
|
||||
+ timedelta(hours=settings.AI_CONVERSATION_TTL_HOURS),
|
||||
)
|
||||
db.add(conversation)
|
||||
await db.flush()
|
||||
return conversation
|
||||
|
||||
|
||||
async def get_conversation(
|
||||
conversation_id: UUID,
|
||||
user_id: UUID,
|
||||
db: AsyncSession,
|
||||
) -> AIConversation:
|
||||
"""Get a conversation, validating ownership and expiry.
|
||||
|
||||
Raises HTTPException 410 if expired, 404 if not found or wrong owner.
|
||||
"""
|
||||
result = await db.execute(
|
||||
select(AIConversation).where(AIConversation.id == conversation_id)
|
||||
)
|
||||
conversation = result.scalar_one_or_none()
|
||||
|
||||
if not conversation or conversation.user_id != user_id:
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_404_NOT_FOUND,
|
||||
detail="Conversation not found",
|
||||
)
|
||||
|
||||
if conversation.expires_at < datetime.now(timezone.utc):
|
||||
conversation.status = "expired"
|
||||
await db.flush()
|
||||
raise HTTPException(
|
||||
status_code=status.HTTP_410_GONE,
|
||||
detail="Conversation expired. Please start a new AI build.",
|
||||
)
|
||||
|
||||
return conversation
|
||||
|
||||
|
||||
async def update_conversation(
|
||||
conversation_id: UUID,
|
||||
user_id: UUID,
|
||||
updates: dict[str, Any],
|
||||
db: AsyncSession,
|
||||
) -> AIConversation:
|
||||
"""Update a conversation's fields.
|
||||
|
||||
Validates ownership and expiry before updating.
|
||||
"""
|
||||
conversation = await get_conversation(conversation_id, user_id, db)
|
||||
|
||||
for key, value in updates.items():
|
||||
if hasattr(conversation, key):
|
||||
setattr(conversation, key, value)
|
||||
|
||||
await db.flush()
|
||||
return conversation
|
||||
186
backend/app/core/ai_quota_service.py
Normal file
186
backend/app/core/ai_quota_service.py
Normal file
@@ -0,0 +1,186 @@
|
||||
"""AI generation quota management.
|
||||
|
||||
Enforces monthly and daily limits on AI flow builder usage.
|
||||
Monthly quota consumed only on successful tree assembly (counts_toward_quota=True).
|
||||
Daily limit is an anti-abuse guard consumed on conversation start.
|
||||
"""
|
||||
import calendar
|
||||
from datetime import datetime, timezone, timedelta
|
||||
from typing import Optional
|
||||
from uuid import UUID
|
||||
|
||||
from sqlalchemy import select, func
|
||||
from sqlalchemy.ext.asyncio import AsyncSession
|
||||
|
||||
from app.models.ai_usage import AIUsage
|
||||
from app.models.plan_limits import PlanLimits
|
||||
from app.models.account_limit_override import AccountLimitOverride
|
||||
from app.core.subscriptions import get_account_subscription, get_plan_limits
|
||||
|
||||
|
||||
async def get_user_plan(account_id: Optional[UUID], db: AsyncSession) -> str:
|
||||
"""Get the plan tier for an account."""
|
||||
if not account_id:
|
||||
return "free"
|
||||
sub = await get_account_subscription(account_id, db)
|
||||
if sub is None:
|
||||
return "free"
|
||||
return sub.plan if sub.plan else "free"
|
||||
|
||||
|
||||
async def _get_effective_limits(
|
||||
account_id: UUID, plan: str, db: AsyncSession
|
||||
) -> tuple[Optional[int], Optional[int]]:
|
||||
"""Get effective AI limits (monthly, daily), applying account overrides.
|
||||
|
||||
Returns (monthly_limit, daily_limit). None means unlimited.
|
||||
"""
|
||||
limits = await get_plan_limits(plan, db)
|
||||
monthly = limits.max_ai_builds_per_month if limits else None
|
||||
daily = limits.max_ai_builds_per_24h if limits else None
|
||||
|
||||
# Check for account-level overrides
|
||||
result = await db.execute(
|
||||
select(AccountLimitOverride).where(
|
||||
AccountLimitOverride.account_id == account_id
|
||||
)
|
||||
)
|
||||
override = result.scalar_one_or_none()
|
||||
if override:
|
||||
if override.override_max_ai_builds_per_month is not None:
|
||||
monthly = override.override_max_ai_builds_per_month
|
||||
if override.override_max_ai_builds_per_24h is not None:
|
||||
daily = override.override_max_ai_builds_per_24h
|
||||
|
||||
return monthly, daily
|
||||
|
||||
|
||||
def _get_billing_anchor_month_start(anchor: Optional[datetime]) -> datetime:
|
||||
"""Calculate the start of the current billing month from the anchor date.
|
||||
|
||||
If the anchor is day 15, the billing month runs from the 15th of each month.
|
||||
Falls back to calendar month if anchor is None.
|
||||
"""
|
||||
now = datetime.now(timezone.utc)
|
||||
if not anchor:
|
||||
return now.replace(day=1, hour=0, minute=0, second=0, microsecond=0)
|
||||
|
||||
anchor_day = min(anchor.day, 28) # Clamp to avoid month overflow
|
||||
this_month_anchor = now.replace(
|
||||
day=anchor_day, hour=0, minute=0, second=0, microsecond=0
|
||||
)
|
||||
|
||||
if now >= this_month_anchor:
|
||||
return this_month_anchor
|
||||
else:
|
||||
# We're before the anchor day, so billing month started last month
|
||||
if now.month == 1:
|
||||
return this_month_anchor.replace(year=now.year - 1, month=12)
|
||||
else:
|
||||
return this_month_anchor.replace(month=now.month - 1)
|
||||
|
||||
|
||||
async def check_ai_quota(
|
||||
user_id: UUID,
|
||||
account_id: UUID,
|
||||
db: AsyncSession,
|
||||
billing_anchor: Optional[datetime] = None,
|
||||
) -> tuple[bool, dict]:
|
||||
"""Check if user can make an AI generation.
|
||||
|
||||
Returns (allowed, quota_status_dict).
|
||||
Monthly counts only rows with counts_toward_quota=True.
|
||||
Daily counts only rows with generation_type in ('scaffold', 'branch_detail').
|
||||
"""
|
||||
plan = await get_user_plan(account_id, db)
|
||||
monthly_limit, daily_limit = await _get_effective_limits(account_id, plan, db)
|
||||
|
||||
now = datetime.now(timezone.utc)
|
||||
month_start = _get_billing_anchor_month_start(billing_anchor)
|
||||
day_start = now - timedelta(hours=24)
|
||||
|
||||
# Monthly: count successful quota-consuming records
|
||||
monthly_count = await db.scalar(
|
||||
select(func.count(AIUsage.id)).where(
|
||||
AIUsage.user_id == user_id,
|
||||
AIUsage.counts_toward_quota == True, # noqa: E712
|
||||
AIUsage.created_at >= month_start,
|
||||
)
|
||||
) or 0
|
||||
|
||||
# Daily: count all AI API calls (scaffold + branch_detail) in last 24h
|
||||
daily_count = await db.scalar(
|
||||
select(func.count(AIUsage.id)).where(
|
||||
AIUsage.user_id == user_id,
|
||||
AIUsage.succeeded == True, # noqa: E712
|
||||
AIUsage.generation_type.in_(["scaffold", "branch_detail"]),
|
||||
AIUsage.created_at >= day_start,
|
||||
)
|
||||
) or 0
|
||||
|
||||
allowed = True
|
||||
deny_reason = None
|
||||
if monthly_limit is not None and monthly_count >= monthly_limit:
|
||||
allowed = False
|
||||
deny_reason = "monthly"
|
||||
if daily_limit is not None and daily_count >= daily_limit:
|
||||
allowed = False
|
||||
deny_reason = "daily"
|
||||
|
||||
# Calculate reset timestamps
|
||||
next_month = month_start.month % 12 + 1
|
||||
next_year = month_start.year + (1 if month_start.month == 12 else 0)
|
||||
max_day = calendar.monthrange(next_year, next_month)[1]
|
||||
monthly_reset_at = month_start.replace(
|
||||
month=next_month,
|
||||
year=next_year,
|
||||
day=min(month_start.day, max_day),
|
||||
)
|
||||
daily_reset_at = day_start + timedelta(hours=24)
|
||||
|
||||
return allowed, {
|
||||
"plan": plan,
|
||||
"monthly_used": monthly_count,
|
||||
"monthly_limit": monthly_limit,
|
||||
"monthly_reset_at": monthly_reset_at.isoformat(),
|
||||
"daily_used": daily_count,
|
||||
"daily_limit": daily_limit,
|
||||
"daily_reset_at": daily_reset_at.isoformat(),
|
||||
"allowed": allowed,
|
||||
"deny_reason": deny_reason,
|
||||
}
|
||||
|
||||
|
||||
async def record_ai_usage(
|
||||
user_id: UUID,
|
||||
account_id: UUID,
|
||||
conversation_id: Optional[UUID],
|
||||
generation_type: str,
|
||||
tier: str,
|
||||
input_tokens: int,
|
||||
output_tokens: int,
|
||||
estimated_cost: float,
|
||||
succeeded: bool,
|
||||
counts_toward_quota: bool,
|
||||
error_code: Optional[str],
|
||||
extra_data: Optional[dict],
|
||||
db: AsyncSession,
|
||||
) -> AIUsage:
|
||||
"""Record an AI usage entry."""
|
||||
usage = AIUsage(
|
||||
user_id=user_id,
|
||||
account_id=account_id,
|
||||
conversation_id=conversation_id,
|
||||
generation_type=generation_type,
|
||||
tier_at_time=tier,
|
||||
input_tokens=input_tokens,
|
||||
output_tokens=output_tokens,
|
||||
estimated_cost_usd=estimated_cost,
|
||||
succeeded=succeeded,
|
||||
counts_toward_quota=counts_toward_quota,
|
||||
error_code=error_code,
|
||||
extra_data=extra_data or {},
|
||||
)
|
||||
db.add(usage)
|
||||
await db.flush()
|
||||
return usage
|
||||
322
backend/app/core/ai_tree_generator_service.py
Normal file
322
backend/app/core/ai_tree_generator_service.py
Normal file
@@ -0,0 +1,322 @@
|
||||
"""AI-powered tree generation service using Anthropic Claude API.
|
||||
|
||||
Implements the 4-stage wizard flow:
|
||||
Stage 2 (scaffold): AI suggests 4-7 top-level branches
|
||||
Stage 3 (branch_detail): AI generates detailed nodes per branch
|
||||
Stage 4 (assemble): Pure assembly logic — zero AI calls
|
||||
|
||||
System prompts are static constants to enable Anthropic prompt caching.
|
||||
"""
|
||||
import json
|
||||
import logging
|
||||
import re
|
||||
import uuid
|
||||
from typing import Any
|
||||
|
||||
import anthropic
|
||||
|
||||
from app.core.config import settings
|
||||
from app.core.ai_tree_validator import validate_generated_tree, count_tree_stats
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
|
||||
# ── Cost estimation (Haiku 4.5 pricing) ──
|
||||
COST_PER_INPUT_TOKEN = 1.0 / 1_000_000 # $1.00 per 1M input tokens
|
||||
COST_PER_OUTPUT_TOKEN = 5.0 / 1_000_000 # $5.00 per 1M output tokens
|
||||
|
||||
|
||||
# ── System Prompts ──
|
||||
|
||||
SCAFFOLD_SYSTEM_PROMPT = """You are ResolutionFlow AI, assisting MSP engineers to build troubleshooting and procedural flows for IT service management.
|
||||
|
||||
Context: Your audience is technical MSP staff experienced with Windows Server, Active Directory, networking, and common MSP tooling (ConnectWise, Datto, SonicWall, etc.).
|
||||
|
||||
Task: Given a flow type, category, name, description, and environment tags, suggest 4-7 top-level branches for the flow.
|
||||
|
||||
For TROUBLESHOOTING flows:
|
||||
- Branches should describe the symptom the user observes — written as what the user sees or reports
|
||||
- The branch name becomes a selectable option on the first screen, so it must be self-identifying from a user's perspective
|
||||
- Good: "Drive letter missing after login", "Mapped drive shows as disconnected (red X)", "Access denied when opening files"
|
||||
- Bad: "Authentication Failures", "GPO Issues", "Connectivity Problems" — too vague for users to self-identify
|
||||
- Order from most common to least common
|
||||
|
||||
For PROCEDURE flows:
|
||||
- Branches should be phase-based stages (e.g., "Prerequisites", "Configuration", "Verification", "Documentation")
|
||||
- Each branch represents a major step in the process
|
||||
- Order in logical execution sequence
|
||||
|
||||
Rules:
|
||||
- Suggest 4-7 branches
|
||||
- Be specific to the technology/service described — avoid generic internal category names
|
||||
- Branch names should be concise (3-7 words) and describe the observable symptom or phase
|
||||
- Each branch needs a brief description (1 sentence) explaining what scenarios it covers
|
||||
- Return ONLY valid JSON, no markdown, no explanation
|
||||
|
||||
Output format:
|
||||
{"branches": [{"name": "Branch Name", "description": "Brief description of what this covers"}]}"""
|
||||
|
||||
|
||||
BRANCH_DETAIL_SYSTEM_PROMPT = """You are ResolutionFlow AI generating step-by-step detail for one branch of a troubleshooting or procedural flow for MSP engineers.
|
||||
|
||||
Context: Your audience is technical MSP staff experienced with Windows Server, Active Directory, networking, and common MSP tooling.
|
||||
|
||||
You must return ONLY valid JSON — no markdown, no code fences, no explanation.
|
||||
|
||||
Required node schema:
|
||||
|
||||
Decision nodes (branching diagnostic questions — choose the right path):
|
||||
{"id": "unique-slug", "type": "decision", "question": "The diagnostic question", "help_text": "Optional context or command hint", "options": [{"id": "opt-id", "label": "Specific observable answer", "next_node_id": "child-node-id"}], "children": [<all child nodes listed here>]}
|
||||
|
||||
Action nodes (a single investigation or remediation step — MUST have next_node_id pointing to the next node):
|
||||
{"id": "unique-slug", "type": "action", "title": "Short title", "description": "Detailed instructions", "commands": ["PowerShell or CMD commands"], "expected_outcome": "What success looks like", "next_node_id": "id-of-next-sibling-node"}
|
||||
|
||||
Solution nodes (leaf nodes — the final resolution, no children):
|
||||
{"id": "unique-slug", "type": "solution", "title": "Resolution title", "description": "Full resolution description", "resolution_steps": ["Step 1", "Step 2"]}
|
||||
|
||||
CRITICAL NAVIGATION RULES:
|
||||
- Decision node: each option's next_node_id MUST exactly match the "id" of a direct child in that decision node's "children" array
|
||||
- Action node: next_node_id MUST exactly match the "id" of a sibling node (another child of the same parent decision node)
|
||||
- Every action node MUST have a next_node_id — action nodes with no next step are broken dead ends
|
||||
- Solution nodes have no children and no next_node_id — they are the terminus
|
||||
- Every path through the tree MUST end at a solution node
|
||||
|
||||
Additional rules:
|
||||
1. Generate 4-10 nodes total for this branch
|
||||
2. Start with a decision node if troubleshooting, action node if procedure
|
||||
3. Decision nodes must have at least 2 options with specific, observable answer choices
|
||||
4. Include realistic MSP commands (PowerShell preferred for Windows)
|
||||
5. Use unique node IDs prefixed with the branch context (e.g., "gpo-check-link")
|
||||
6. Build the tree bottom-up in your head: create solution/leaf nodes first, then build parent nodes referencing their IDs
|
||||
|
||||
Few-shot example showing correct action node next_node_id usage:
|
||||
{"id": "dns-root", "type": "decision", "question": "Can the client resolve any DNS names?", "help_text": "Run: nslookup google.com", "options": [{"id": "dns-opt-none", "label": "No — nslookup times out or returns 'server failed'", "next_node_id": "dns-check-service"}, {"id": "dns-opt-partial", "label": "Some names resolve but others fail", "next_node_id": "dns-check-specific"}], "children": [{"id": "dns-check-service", "type": "action", "title": "Check DNS Client Service", "description": "Verify the DNS Client service is running on the affected machine", "commands": ["Get-Service -Name Dnscache | Select-Object Status,StartType"], "expected_outcome": "Status should be Running", "next_node_id": "dns-service-solution"}, {"id": "dns-service-solution", "type": "solution", "title": "DNS Service Was Stopped", "description": "The DNS Client service was stopped, preventing all name resolution", "resolution_steps": ["Run: Start-Service Dnscache", "Set startup type: Set-Service Dnscache -StartupType Automatic", "Flush cache: ipconfig /flushdns", "Test: nslookup google.com"]}, {"id": "dns-check-specific", "type": "solution", "title": "Selective DNS Failure — Stale or Missing Records", "description": "Some records resolve correctly, indicating DNS is functional but specific records are stale or missing", "resolution_steps": ["Check DNS server for missing A/CNAME records", "Clear DNS cache on the DNS server: Clear-DnsServerCache", "Flush client cache: ipconfig /flushdns", "Verify with: nslookup <failing-hostname>"]}]}"""
|
||||
|
||||
|
||||
CORRECTIVE_PROMPT_TEMPLATE = """Your previous JSON was invalid for ResolutionFlow's tree schema.
|
||||
|
||||
Validation errors:
|
||||
{error_list}
|
||||
|
||||
IMPORTANT: If any error mentions a next_node_id referencing a non-existent child, you must ensure every option's next_node_id exactly matches the "id" field of one of the node's direct children. The child node must exist in the "children" array of the same parent node.
|
||||
|
||||
Return a corrected full JSON object only. No markdown, no prose, no code fences.
|
||||
Fix ALL listed errors while maintaining the same troubleshooting/procedural logic."""
|
||||
|
||||
|
||||
def _strip_markdown_fences(text: str) -> str:
|
||||
"""Strip markdown code fences if the model wrapped its JSON response."""
|
||||
text = text.strip()
|
||||
match = re.match(r"^```(?:json)?\s*([\s\S]*?)```$", text)
|
||||
if match:
|
||||
return match.group(1).strip()
|
||||
return text
|
||||
|
||||
|
||||
def _get_client() -> anthropic.AsyncAnthropic:
|
||||
"""Get configured async Anthropic client."""
|
||||
if not settings.ANTHROPIC_API_KEY:
|
||||
raise RuntimeError("ANTHROPIC_API_KEY not configured")
|
||||
return anthropic.AsyncAnthropic(
|
||||
api_key=settings.ANTHROPIC_API_KEY,
|
||||
timeout=settings.AI_REQUEST_TIMEOUT_SECONDS,
|
||||
)
|
||||
|
||||
|
||||
def _estimate_cost(input_tokens: int, output_tokens: int) -> float:
|
||||
"""Estimate USD cost from token counts."""
|
||||
return (input_tokens * COST_PER_INPUT_TOKEN) + (
|
||||
output_tokens * COST_PER_OUTPUT_TOKEN
|
||||
)
|
||||
|
||||
|
||||
async def scaffold_branches(
|
||||
wizard_state: dict[str, Any],
|
||||
) -> tuple[list[dict[str, str]], int, int, float]:
|
||||
"""Stage 2: AI suggests top-level branches.
|
||||
|
||||
Returns (branches, input_tokens, output_tokens, estimated_cost).
|
||||
Raises ValueError on invalid response.
|
||||
"""
|
||||
client = _get_client()
|
||||
|
||||
flow_type = wizard_state.get("flow_type", "troubleshooting")
|
||||
name = wizard_state.get("name", "")
|
||||
description = wizard_state.get("description", "")
|
||||
tags = wizard_state.get("environment_tags", [])
|
||||
|
||||
user_message = (
|
||||
f"Flow type: {flow_type}\n"
|
||||
f"Name: {name}\n"
|
||||
f"Description: {description}\n"
|
||||
)
|
||||
if tags:
|
||||
user_message += f"Environment: {', '.join(tags)}\n"
|
||||
|
||||
response = await client.messages.create(
|
||||
model=settings.AI_MODEL,
|
||||
max_tokens=1024,
|
||||
system=SCAFFOLD_SYSTEM_PROMPT,
|
||||
messages=[{"role": "user", "content": user_message}],
|
||||
)
|
||||
|
||||
raw_text = _strip_markdown_fences(response.content[0].text)
|
||||
input_tokens = response.usage.input_tokens
|
||||
output_tokens = response.usage.output_tokens
|
||||
cost = _estimate_cost(input_tokens, output_tokens)
|
||||
|
||||
try:
|
||||
data = json.loads(raw_text)
|
||||
except json.JSONDecodeError as e:
|
||||
raise ValueError(f"AI returned invalid JSON: {e}")
|
||||
|
||||
branches = data.get("branches", [])
|
||||
if not isinstance(branches, list) or len(branches) < 2:
|
||||
raise ValueError("AI returned fewer than 2 branches")
|
||||
|
||||
return branches, input_tokens, output_tokens, cost
|
||||
|
||||
|
||||
async def generate_branch_detail(
|
||||
wizard_state: dict[str, Any],
|
||||
branch_name: str,
|
||||
existing_branches: list[str],
|
||||
) -> tuple[dict[str, Any], int, int, float]:
|
||||
"""Stage 3: AI generates detailed nodes for one branch.
|
||||
|
||||
Returns (branch_tree, input_tokens, output_tokens, estimated_cost).
|
||||
On validation failure, retries once with corrective prompt.
|
||||
Raises ValueError if both attempts fail.
|
||||
"""
|
||||
client = _get_client()
|
||||
|
||||
flow_type = wizard_state.get("flow_type", "troubleshooting")
|
||||
name = wizard_state.get("name", "")
|
||||
description = wizard_state.get("description", "")
|
||||
|
||||
user_message = (
|
||||
f"Flow: {name} ({flow_type})\n"
|
||||
f"Description: {description}\n"
|
||||
f"Branch to detail: {branch_name}\n"
|
||||
)
|
||||
if existing_branches:
|
||||
other = [b for b in existing_branches if b != branch_name]
|
||||
if other:
|
||||
user_message += f"Other branches (avoid overlap): {', '.join(other)}\n"
|
||||
|
||||
messages = [{"role": "user", "content": user_message}]
|
||||
total_input = 0
|
||||
total_output = 0
|
||||
|
||||
for attempt in range(3):
|
||||
response = await client.messages.create(
|
||||
model=settings.AI_MODEL,
|
||||
max_tokens=8192,
|
||||
system=BRANCH_DETAIL_SYSTEM_PROMPT,
|
||||
messages=messages,
|
||||
)
|
||||
|
||||
total_input += response.usage.input_tokens
|
||||
total_output += response.usage.output_tokens
|
||||
logger.debug(
|
||||
"branch_detail attempt=%d stop_reason=%s content_blocks=%d output_tokens=%d",
|
||||
attempt,
|
||||
response.stop_reason,
|
||||
len(response.content),
|
||||
response.usage.output_tokens,
|
||||
)
|
||||
raw_text = _strip_markdown_fences(response.content[0].text) if response.content else ""
|
||||
if not raw_text:
|
||||
logger.warning("branch_detail attempt=%d returned empty text, stop_reason=%s", attempt, response.stop_reason)
|
||||
|
||||
try:
|
||||
branch_tree = json.loads(raw_text)
|
||||
except json.JSONDecodeError as e:
|
||||
if attempt < 2:
|
||||
messages.append({"role": "assistant", "content": raw_text})
|
||||
messages.append({
|
||||
"role": "user",
|
||||
"content": CORRECTIVE_PROMPT_TEMPLATE.format(
|
||||
error_list=f"JSON parse error: {e}"
|
||||
),
|
||||
})
|
||||
continue
|
||||
raise ValueError(f"AI returned invalid JSON after retry: {e}")
|
||||
|
||||
errors = validate_generated_tree(branch_tree)
|
||||
if not errors:
|
||||
cost = _estimate_cost(total_input, total_output)
|
||||
return branch_tree, total_input, total_output, cost
|
||||
|
||||
if attempt < 2:
|
||||
messages.append({"role": "assistant", "content": raw_text})
|
||||
messages.append({
|
||||
"role": "user",
|
||||
"content": CORRECTIVE_PROMPT_TEMPLATE.format(
|
||||
error_list="\n".join(f"- {e}" for e in errors)
|
||||
),
|
||||
})
|
||||
continue
|
||||
|
||||
raise ValueError(
|
||||
f"AI tree validation failed after retry: {'; '.join(errors)}"
|
||||
)
|
||||
|
||||
# Should not reach here
|
||||
raise ValueError("Branch detail generation failed")
|
||||
|
||||
|
||||
def assemble_tree(
|
||||
wizard_state: dict[str, Any],
|
||||
branches: list[dict[str, Any]],
|
||||
) -> tuple[dict[str, Any], str, str, dict[str, int]]:
|
||||
"""Stage 4: Assemble branches into a complete tree.
|
||||
|
||||
Zero AI calls — pure assembly logic.
|
||||
Returns (tree_structure, suggested_name, suggested_description, summary_stats).
|
||||
"""
|
||||
flow_type = wizard_state.get("flow_type", "troubleshooting")
|
||||
name = wizard_state.get("name", "Untitled Flow")
|
||||
description = wizard_state.get("description", "")
|
||||
|
||||
# Build root decision node pointing to each branch
|
||||
options = []
|
||||
children = []
|
||||
for i, branch in enumerate(branches):
|
||||
branch_name = branch.get("name", f"Branch {i + 1}")
|
||||
branch_tree = branch.get("steps")
|
||||
|
||||
if not branch_tree or not isinstance(branch_tree, dict):
|
||||
# Skip branches without detail
|
||||
continue
|
||||
|
||||
branch_id = branch_tree.get("id", f"branch_{i}")
|
||||
options.append({
|
||||
"id": f"opt_{i + 1}",
|
||||
"label": branch_name,
|
||||
"next_node_id": branch_id,
|
||||
})
|
||||
children.append(branch_tree)
|
||||
|
||||
if len(options) < 2:
|
||||
raise ValueError("Need at least 2 branches with detail to assemble a tree")
|
||||
|
||||
# Determine root question based on flow type
|
||||
if flow_type == "troubleshooting":
|
||||
root_question = f"What is the user experiencing? Select the symptom that best matches their report."
|
||||
root_help = "Choose the option that most closely describes the user's reported problem."
|
||||
else:
|
||||
root_question = f"Which phase of {name} are you working on?"
|
||||
root_help = None
|
||||
|
||||
tree_structure = {
|
||||
"id": "root",
|
||||
"type": "decision",
|
||||
"question": root_question,
|
||||
**({"help_text": root_help} if root_help else {}),
|
||||
"options": options,
|
||||
"children": children,
|
||||
}
|
||||
|
||||
stats = count_tree_stats(tree_structure)
|
||||
|
||||
return tree_structure, name, description, stats
|
||||
217
backend/app/core/ai_tree_validator.py
Normal file
217
backend/app/core/ai_tree_validator.py
Normal file
@@ -0,0 +1,217 @@
|
||||
"""Validation for AI-generated tree structures.
|
||||
|
||||
Ensures generated trees conform to ResolutionFlow's node schema
|
||||
before they are saved to the database.
|
||||
"""
|
||||
from typing import Any
|
||||
|
||||
|
||||
VALID_NODE_TYPES = {"decision", "action", "solution"}
|
||||
|
||||
# Required fields per node type
|
||||
REQUIRED_FIELDS = {
|
||||
"decision": {"id", "type", "question", "options", "children"},
|
||||
"action": {"id", "type", "title", "description"},
|
||||
"solution": {"id", "type", "title", "description"},
|
||||
}
|
||||
|
||||
|
||||
class TreeValidationError(Exception):
|
||||
"""Raised when a generated tree fails validation."""
|
||||
|
||||
def __init__(self, errors: list[str]):
|
||||
self.errors = errors
|
||||
super().__init__(f"Tree validation failed: {'; '.join(errors)}")
|
||||
|
||||
|
||||
def validate_generated_tree(tree: dict[str, Any]) -> list[str]:
|
||||
"""Validate an AI-generated tree structure.
|
||||
|
||||
Returns a list of error strings. Empty list means valid.
|
||||
"""
|
||||
errors: list[str] = []
|
||||
|
||||
if not isinstance(tree, dict):
|
||||
return ["Tree must be a JSON object"]
|
||||
|
||||
# Root must be a decision node
|
||||
if tree.get("type") != "decision":
|
||||
errors.append("Root node must be type 'decision'")
|
||||
|
||||
# Collect all node IDs and validate structure
|
||||
all_ids: set[str] = set()
|
||||
all_referenced_ids: set[str] = set()
|
||||
node_count = 0
|
||||
solution_count = 0
|
||||
|
||||
def _validate_node(node: dict[str, Any], path: str) -> None:
|
||||
nonlocal node_count, solution_count
|
||||
|
||||
if not isinstance(node, dict):
|
||||
errors.append(f"Node at {path} is not an object")
|
||||
return
|
||||
|
||||
node_count += 1
|
||||
node_type = node.get("type")
|
||||
node_id = node.get("id")
|
||||
|
||||
# Check node ID
|
||||
if not node_id:
|
||||
errors.append(f"Node at {path} missing 'id'")
|
||||
elif node_id in all_ids:
|
||||
errors.append(f"Duplicate node ID: '{node_id}'")
|
||||
else:
|
||||
all_ids.add(node_id)
|
||||
|
||||
# Check node type
|
||||
if node_type not in VALID_NODE_TYPES:
|
||||
errors.append(
|
||||
f"Node '{node_id or path}' has invalid type '{node_type}'. "
|
||||
f"Must be one of: {', '.join(sorted(VALID_NODE_TYPES))}"
|
||||
)
|
||||
return
|
||||
|
||||
# Check required fields
|
||||
required = REQUIRED_FIELDS[node_type]
|
||||
missing = required - set(node.keys())
|
||||
if missing:
|
||||
errors.append(
|
||||
f"Node '{node_id}' (type={node_type}) missing fields: {', '.join(sorted(missing))}"
|
||||
)
|
||||
|
||||
# Type-specific validation
|
||||
if node_type == "decision":
|
||||
options = node.get("options", [])
|
||||
if not isinstance(options, list) or len(options) < 2:
|
||||
errors.append(
|
||||
f"Decision node '{node_id}' must have at least 2 options"
|
||||
)
|
||||
else:
|
||||
children = node.get("children", [])
|
||||
child_ids = {c.get("id") for c in children if isinstance(c, dict)}
|
||||
option_ids: set[str] = set()
|
||||
|
||||
for opt in options:
|
||||
if not isinstance(opt, dict):
|
||||
errors.append(f"Option in node '{node_id}' is not an object")
|
||||
continue
|
||||
opt_id = opt.get("id")
|
||||
if opt_id and opt_id in option_ids:
|
||||
errors.append(
|
||||
f"Duplicate option ID '{opt_id}' in node '{node_id}'"
|
||||
)
|
||||
if opt_id:
|
||||
option_ids.add(opt_id)
|
||||
|
||||
next_id = opt.get("next_node_id")
|
||||
if next_id:
|
||||
all_referenced_ids.add(next_id)
|
||||
if child_ids and next_id not in child_ids:
|
||||
errors.append(
|
||||
f"Option '{opt.get('label', '?')}' in node '{node_id}' "
|
||||
f"references non-existent child '{next_id}'"
|
||||
)
|
||||
|
||||
elif node_type == "action":
|
||||
next_id = node.get("next_node_id")
|
||||
if not next_id:
|
||||
errors.append(
|
||||
f"Action node '{node_id}' is missing 'next_node_id'. "
|
||||
"Every action node must point to the next node (a sibling in the parent's children)."
|
||||
)
|
||||
else:
|
||||
all_referenced_ids.add(next_id)
|
||||
|
||||
elif node_type == "solution":
|
||||
solution_count += 1
|
||||
|
||||
# Recurse into children
|
||||
for i, child in enumerate(node.get("children", [])):
|
||||
_validate_node(child, f"{path}.children[{i}]")
|
||||
|
||||
_validate_node(tree, "root")
|
||||
|
||||
# Global checks
|
||||
if node_count < 5:
|
||||
errors.append(
|
||||
f"Tree has only {node_count} nodes. Minimum 5 required for a useful tree."
|
||||
)
|
||||
if node_count > 50:
|
||||
errors.append(
|
||||
f"Tree has {node_count} nodes. Maximum 50 allowed."
|
||||
)
|
||||
if solution_count < 2:
|
||||
errors.append(
|
||||
f"Tree has only {solution_count} solution nodes. "
|
||||
"Need at least 2 to cover different resolution paths."
|
||||
)
|
||||
|
||||
# Check that all leaf (non-solution) nodes have children or are solutions
|
||||
_check_branch_termination(tree, errors)
|
||||
|
||||
return errors
|
||||
|
||||
|
||||
def _check_branch_termination(node: dict[str, Any], errors: list[str]) -> None:
|
||||
"""Verify every branch eventually reaches a solution node.
|
||||
|
||||
Action nodes continue via next_node_id (validated separately).
|
||||
Only decision nodes with no children are dead ends.
|
||||
"""
|
||||
if not isinstance(node, dict):
|
||||
return
|
||||
|
||||
node_type = node.get("type")
|
||||
node_id = node.get("id", "?")
|
||||
children = node.get("children", [])
|
||||
|
||||
if node_type == "solution":
|
||||
return # Solution is a valid terminus
|
||||
|
||||
if node_type == "action":
|
||||
# Action nodes continue via next_node_id (a sibling), not children.
|
||||
# next_node_id presence is validated in _validate_node.
|
||||
# Recurse into children if present (non-standard but tolerate it).
|
||||
for child in children:
|
||||
_check_branch_termination(child, errors)
|
||||
return
|
||||
|
||||
# Decision node: must have children
|
||||
if not children:
|
||||
errors.append(
|
||||
f"Decision node '{node_id}' is a dead end — "
|
||||
"it has no children"
|
||||
)
|
||||
return
|
||||
|
||||
for child in children:
|
||||
_check_branch_termination(child, errors)
|
||||
|
||||
|
||||
def count_tree_stats(tree: dict[str, Any]) -> dict[str, int]:
|
||||
"""Count node types and calculate depth of a tree."""
|
||||
stats = {
|
||||
"node_count": 0,
|
||||
"decision_count": 0,
|
||||
"action_count": 0,
|
||||
"solution_count": 0,
|
||||
"depth": 0,
|
||||
}
|
||||
|
||||
def _count(node: dict[str, Any], depth: int) -> None:
|
||||
if not isinstance(node, dict):
|
||||
return
|
||||
stats["node_count"] += 1
|
||||
node_type = node.get("type", "")
|
||||
if node_type == "decision":
|
||||
stats["decision_count"] += 1
|
||||
elif node_type == "action":
|
||||
stats["action_count"] += 1
|
||||
elif node_type == "solution":
|
||||
stats["solution_count"] += 1
|
||||
stats["depth"] = max(stats["depth"], depth)
|
||||
for child in node.get("children", []):
|
||||
_count(child, depth + 1)
|
||||
|
||||
_count(tree, 1)
|
||||
return stats
|
||||
@@ -72,6 +72,18 @@ class Settings(BaseSettings):
|
||||
"""Check if Stripe is configured."""
|
||||
return self.STRIPE_SECRET_KEY is not None and self.STRIPE_WEBHOOK_SECRET is not None
|
||||
|
||||
# AI Flow Builder
|
||||
ANTHROPIC_API_KEY: Optional[str] = None
|
||||
AI_MODEL: str = "claude-haiku-4-5-20251001"
|
||||
AI_CONVERSATION_TTL_HOURS: int = 24
|
||||
AI_MAX_CALLS_PER_FLOW: int = 10
|
||||
AI_REQUEST_TIMEOUT_SECONDS: int = 45
|
||||
|
||||
@property
|
||||
def ai_enabled(self) -> bool:
|
||||
"""Check if AI Flow Builder is configured."""
|
||||
return self.ANTHROPIC_API_KEY is not None
|
||||
|
||||
# Deployment – auto-seed test data on PR environments
|
||||
SEED_ON_DEPLOY: bool = False
|
||||
|
||||
|
||||
@@ -1,4 +1,4 @@
|
||||
"""APScheduler integration for maintenance flow auto-session creation."""
|
||||
"""APScheduler integration for maintenance flow auto-session creation and AI cleanup."""
|
||||
import logging
|
||||
import uuid
|
||||
from datetime import datetime, timezone
|
||||
@@ -7,8 +7,9 @@ from apscheduler.schedulers.asyncio import AsyncIOScheduler
|
||||
from apscheduler.schedulers.base import SchedulerNotRunningError
|
||||
from apscheduler.jobstores.base import JobLookupError
|
||||
from apscheduler.triggers.cron import CronTrigger
|
||||
from apscheduler.triggers.interval import IntervalTrigger
|
||||
import pytz
|
||||
from sqlalchemy import select
|
||||
from sqlalchemy import select, delete
|
||||
from sqlalchemy.ext.asyncio import AsyncSession
|
||||
|
||||
logger = logging.getLogger(__name__)
|
||||
@@ -114,6 +115,27 @@ async def _fire_maintenance_schedule(schedule_id: str) -> None:
|
||||
await db.rollback()
|
||||
|
||||
|
||||
async def _cleanup_expired_ai_conversations() -> None:
|
||||
"""Delete expired AI wizard conversations."""
|
||||
import app.models # noqa: F401
|
||||
from app.core.database import async_session_maker
|
||||
from app.models.ai_conversation import AIConversation
|
||||
|
||||
async with async_session_maker() as db:
|
||||
try:
|
||||
result = await db.execute(
|
||||
delete(AIConversation).where(
|
||||
AIConversation.expires_at < datetime.now(timezone.utc)
|
||||
)
|
||||
)
|
||||
if result.rowcount > 0:
|
||||
logger.info(f"Cleaned up {result.rowcount} expired AI conversation(s)")
|
||||
await db.commit()
|
||||
except Exception:
|
||||
logger.exception("Error cleaning up expired AI conversations")
|
||||
await db.rollback()
|
||||
|
||||
|
||||
async def load_all_schedules(db: AsyncSession) -> None:
|
||||
"""Load all active schedules into APScheduler on startup."""
|
||||
# Import all models to ensure SQLAlchemy mapper relationships resolve
|
||||
|
||||
@@ -28,7 +28,7 @@ def convert_session_to_tree(
|
||||
return {
|
||||
"id": str(uuid.uuid4()),
|
||||
"type": "solution",
|
||||
"solution": "Session had no recorded path",
|
||||
"title": "Session had no recorded path",
|
||||
"children": []
|
||||
}
|
||||
|
||||
@@ -63,7 +63,7 @@ def convert_session_to_tree(
|
||||
new_node = {
|
||||
"id": node_id,
|
||||
"type": "action",
|
||||
"action": f"Step from original tree (node {node_id})",
|
||||
"title": f"Step from original tree (node {node_id})",
|
||||
"children": []
|
||||
}
|
||||
|
||||
@@ -130,15 +130,15 @@ def _create_node_from_original(
|
||||
if decision and decision.get("answer"):
|
||||
new_node["question"] += f"\n\nAnswer: {decision['answer']}"
|
||||
elif node_type == "action":
|
||||
new_node["action"] = original_node.get("action", "")
|
||||
new_node["title"] = original_node.get("title", original_node.get("action", ""))
|
||||
if decision and decision.get("action_performed"):
|
||||
new_node["action"] = decision["action_performed"]
|
||||
new_node["title"] = decision["action_performed"]
|
||||
if decision and decision.get("command_output"):
|
||||
output = decision["command_output"].strip()
|
||||
if output:
|
||||
new_node["action"] += f"\n\nCommand Output:\n{output}"
|
||||
new_node["title"] += f"\n\nCommand Output:\n{output}"
|
||||
elif node_type == "solution":
|
||||
new_node["solution"] = original_node.get("solution", "")
|
||||
new_node["title"] = original_node.get("title", original_node.get("solution", ""))
|
||||
|
||||
return new_node
|
||||
|
||||
@@ -169,18 +169,18 @@ def _create_node_from_custom_step(
|
||||
if step_type == "decision":
|
||||
new_node["question"] = content
|
||||
elif step_type == "action":
|
||||
new_node["action"] = content
|
||||
new_node["title"] = content
|
||||
elif step_type == "solution":
|
||||
new_node["solution"] = content
|
||||
new_node["title"] = content
|
||||
|
||||
# Add notes if present
|
||||
if custom_step.get("notes"):
|
||||
if step_type == "decision":
|
||||
new_node["question"] += f"\n\nNotes: {custom_step['notes']}"
|
||||
elif step_type == "action":
|
||||
new_node["action"] += f"\n\nNotes: {custom_step['notes']}"
|
||||
new_node["title"] += f"\n\nNotes: {custom_step['notes']}"
|
||||
elif step_type == "solution":
|
||||
new_node["solution"] += f"\n\nNotes: {custom_step['notes']}"
|
||||
new_node["title"] += f"\n\nNotes: {custom_step['notes']}"
|
||||
|
||||
return new_node
|
||||
|
||||
|
||||
@@ -83,17 +83,17 @@ def _validate_node(node: dict[str, Any], path: str, errors: list[dict[str, str]]
|
||||
})
|
||||
|
||||
elif node_type == "action":
|
||||
if "action" not in node or not node["action"]:
|
||||
if "title" not in node or not node["title"]:
|
||||
errors.append({
|
||||
"field": f"{path}.action",
|
||||
"message": "Action nodes must have a non-empty action"
|
||||
"field": f"{path}.title",
|
||||
"message": "Action nodes must have a non-empty title"
|
||||
})
|
||||
|
||||
elif node_type == "solution":
|
||||
if "solution" not in node or not node["solution"]:
|
||||
if "title" not in node or not node["title"]:
|
||||
errors.append({
|
||||
"field": f"{path}.solution",
|
||||
"message": "Solution nodes must have a non-empty solution"
|
||||
"field": f"{path}.title",
|
||||
"message": "Solution nodes must have a non-empty title"
|
||||
})
|
||||
|
||||
elif node_type == "answer":
|
||||
|
||||
Reference in New Issue
Block a user