Michael Chihlas
d0ebdef9e8
All checks were successful
Mirror to GitHub / mirror (push) Successful in 10s
The "AI parrots example content from system prompt" bug bit us twice in one day across two different prompt sites. Patching individual prompts is treating the symptom; this commit makes the rule structural. Audit + sanitize: - assistant_chat_service.ASSISTANT_SYSTEM_PROMPT — already cleaned in prior commits, but the [FORK] schema still had literal "Brief reason" / "Short name" / "One sentence" placeholders. Replaced with <angle-bracket> placeholders. Anti-parrot rule itself rewritten to describe the failure mode abstractly instead of naming "jsmith" so the rule no longer trips the guardrail (and so the model doesn't see "jsmith" as a token at all). - ai_chat_service.py — removed three concrete-example offenders: "Get-Service ADSync" command literal, the "DC01 server_name" intake form payload (in two places), and the inline interview demos using "Azure AD Sync failures" / "Exchange Online mailbox migration". Replaced with technology-neutral schema descriptions. - ai_tree_generator_service.BRANCH_DETAIL_SYSTEM_PROMPT — replaced the fully-fleshed DNS troubleshooting tree (with literal Dnscache / ipconfig / google.com / Start-Service) with a placeholder schema showing only ID-linkage shape. - kb_conversion_service.PROCEDURAL_SYSTEM_PROMPT — replaced the worked Server Manager + DC01 example payload with a placeholder schema. Guardrail (tests/test_prompt_anti_parrot.py): - Imports every module under app/services/ and app/core/ and walks every uppercase string constant ending in _PROMPT, _SCHEMA, _PROTOCOL, _FORMAT, or _CONTEXT. - test 1: known-leaked-token list (jsmith, DC01, ADSync, Dnscache, google.com, "Outlook keeps", "Teams drops") must not appear in any prompt constant. Add to the list when a new leak shows up in prod — the list IS the audit trail. - test 2: marker blocks ([QUESTIONS], [ACTIONS], [SUGGEST_FIX], etc.) must contain placeholders only. Distinguishes JSON keys (followed by ':', allowed) from JSON values (followed by ',' / ']' / '}', must be <placeholder>); allows pipe-separated enum types (text|password|select) and a small set of fixed enum values (question, diagnostic_check, decision, action, ...). Verified by feeding the test a known-bad block — caught it correctly. Documented the rule in CLAUDE.md → AI / FlowPilot lessons, naming the test as the enforcement point so future contributors know how to extend it (add to the known-leaked list when a new leak surfaces). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ResolutionFlow
Stop writing ticket notes. Start generating them.
ResolutionFlow is an AI-powered troubleshooting platform for MSP professionals. Engineers follow guided flows while an AI copilot assists — and documentation writes itself as a byproduct of the work.
Production: resolutionflow.com
Quick Start
# Prerequisites: Docker, Python 3.11+, Node.js 20+
# Start PostgreSQL
docker start patherly_postgres
# Backend
cd backend
source venv/bin/activate
pip install -r requirements.txt
alembic upgrade head
uvicorn app.main:app --reload
# Frontend (separate terminal)
cd frontend
npm install
npm run dev
- Frontend: http://localhost:5173
- Backend API: http://localhost:8000
- API Docs: http://localhost:8000/api/docs
See DEV-ENV.md for full environment setup (devserver, Docker, CORS).
Features
FlowPilot AI Copilot
Like having a senior engineer on every call. FlowPilot guides troubleshooting decisions, suggests next steps with context-aware intelligence, and automatically captures documentation as a byproduct of the session.
- Confidence-tiered model routing (fast responses for simple steps, deeper reasoning for complex decisions)
- AI-generated ticket summaries and session documentation
- Standalone assistant chat with RAG for open-ended troubleshooting
- Knowledge Flywheel: AI analyzes completed sessions and proposes new flows automatically
Guided Flows
- Troubleshooting Flows — Decision trees with branching paths for diagnosing issues
- Procedural Flows (Projects) — Step-by-step checklists for onboarding, migrations, deployments
- Maintenance Flows — Scheduled recurring tasks with batch execution across multiple targets
- Visual Flow Editor with drag-and-drop canvas, undo/redo, markdown support
- AI Flow Builder — describe what you need, get a complete flow generated
Auto-Documentation
Every session generates timestamped, detailed notes formatted for your PSA. Engineers never write another ticket note.
- Export to Markdown, plain text, or HTML
- Sensitive data redaction
- One-click push to ConnectWise PSA tickets
ConnectWise PSA Integration
- Post session documentation directly to ConnectWise tickets as internal notes
- Pull ticket details and client context into FlowPilot sessions
- Member mapping between ResolutionFlow and ConnectWise users
- Credentials encrypted at rest (Fernet), stored per-team
Team & Knowledge Management
- Role-based access (super_admin, team_admin, engineer, viewer)
- Shared flow library with categories, tags, folders, full-text search
- Step Library — reusable troubleshooting steps with ratings and reviews
- Session sharing via link (authenticated and public views)
- Escalation workflow with AI-enhanced briefing packages
- Flow proposals from AI analysis (review queue for team leads)
Tech Stack
| Layer | Technology |
|---|---|
| Frontend | React 19, TypeScript, Vite, Tailwind CSS v4 |
| State | Zustand (immer + zundo for undo/redo) |
| Routing | React Router v7 |
| Canvas | @xyflow/react (React Flow) + dagre |
| Backend | Python FastAPI, async SQLAlchemy 2.0 + asyncpg |
| Database | PostgreSQL 16 |
| Migrations | Alembic (75+ migrations) |
| Auth | JWT (python-jose) + bcrypt, refresh token rotation |
| AI | Anthropic Claude API (tiered model routing) |
| Embeddings | Voyage AI (semantic search) |
| Scheduling | APScheduler 3.x (async) |
| Analytics | PostHog |
| Hosting | Railway (auto-deploy on push to main) |
Project Structure
patherly/
├── backend/
│ ├── app/
│ │ ├── main.py # FastAPI entry point
│ │ ├── api/endpoints/ # Route handlers (35+ endpoints)
│ │ ├── core/ # Config, database, permissions, security
│ │ ├── models/ # SQLAlchemy models
│ │ ├── schemas/ # Pydantic schemas
│ │ └── services/psa/ # PSA provider abstraction layer
│ ├── alembic/ # Database migrations
│ └── tests/ # Integration tests (100+)
├── frontend/
│ ├── src/
│ │ ├── components/ # UI components by domain
│ │ ├── pages/ # Page components
│ │ ├── store/ # Zustand stores
│ │ └── types/ # TypeScript interfaces
├── docs/ # Design docs, plans, ConnectWise reference
├── brand-assets/ # SVGs, brand guide
├── CLAUDE.md # AI assistant project context
├── CURRENT-STATE.md # Detailed feature status
└── CHANGELOG.md # Release history
Running Tests
# Backend integration tests
cd backend
pytest --override-ini="addopts="
# Frontend build (stricter than tsc --noEmit)
cd frontend
npm run build
Documentation
| Document | Purpose |
|---|---|
| CLAUDE.md | Full project context for AI-assisted development |
| CURRENT-STATE.md | Detailed feature status |
| 03-DEVELOPMENT-ROADMAP.md | Development roadmap |
| UI-DESIGN-SYSTEM.md | Design system (Slate & Ice) |
| DEV-ENV.md | Development environment setup |
| CHANGELOG.md | Release history |
License
Proprietary. All rights reserved.