feat: Add simplified server + dashboard, improve security

.github/workflows/ci.yml

@@ -0,0 +1,65 @@
+name: CI
+
+on:
+  pull_request:
+    branches: [main, develop]
+  push:
+    branches: [main, develop]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 10
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: '3.12'
+          cache: 'pip'
+
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install -r requirements.txt
+          pip install pytest pytest-cov black flake8
+
+      - name: Check code formatting with Black
+        run: |
+          black --check .
+          black --check tests/
+
+      - name: Lint with flake8
+        run: |
+          flake8 src/ tests/ --max-line-length=100 --extend-ignore=E203,W503
+
+      - name: Run tests with coverage
+        run: |
+          pytest tests/ -v --cov=src/ --cov-report=xml --cov-report=term
+
+      - name: Upload coverage to Codecov
+        uses: codecov/codecov-action@v4
+        with:
+          fail_ci_if_error: false
+
+  docker-test:
+    runs-on: ubuntu-latest
+    timeout-minutes: 15
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Set up Docker Buildx
+        uses: docker/setup-buildx-action@v3
+
+      - name: Build Docker image
+        run: |
+          docker build -t perplexity-api-simple:test .
+
+      - name: Run tests in container
+        run: |
+          docker run --rm perplexity-api-simple:test python -m pytest tests/ -v

.gitignore

@@ -1,8 +1,22 @@
 # Environment and secrets
 .env
+.env.*
 .api_keys.json
+.api_keys.json.backup
+.api_keys.json.tmp
 .mcp.json
 
+# Runtime data (contains user queries/secrets)
+.query_history.json
+.request_log.jsonl
+.session_savings.json
+.model_stats.json
+.analytics.json
+.alerts.json
+.webhooks.json
+.auto_sync_config.json
+.notifyrc
+
 # Personal AI instructions - NEVER commit
 CLAUDE.md
 GEMINI.md

ATLAS_AI_SESSION_SUMMARY.md

@@ -0,0 +1,80 @@
+# Atlas AI Session Summary
+
+## What We Accomplished
+
+### Phase 1: PR Analysis & Understanding
+- Deep code review of PR #2 (534 lines across 5 files)
+- Ran 3 doubt agents that revealed local/remote repo divergence
+- Researched FastMCP, curl_cffi, and Perplexity API patterns
+- Verified PR findings through live testing
+- **Key discovery:** Local `master` and remote `main` have NO common commits — completely different histories
+
+### Phase 2: Security & Repo Sync
+- **Critical security fix:** Updated `.gitignore` to exclude sensitive runtime data files (`.mcp.json`, `.query_history.json`, `.request_log.jsonl`, etc.)
+- Created `v2` branch from local `master`
+- Pushed public improvements to remote: simplified server, dashboard, ops tooling
+- **Kept private:** Atlas Search MCP server, extension v2, cookie automation scripts
+- Created PR #3 (v2 → main) to properly sync the public repo
+- Posted gracious comment on PR #2 explaining the situation
+
+### Phase 3: Atlas AI Product Strategy
+- Defined product positioning: "Turn AI agent research into professional, cited, structured reports"
+- Designed pricing tiers: Free ($0), Pro ($29/mo), Team ($299/mo), Enterprise (custom)
+- Created path to $10M ARR: 2,800 Pro users + 1,000 Team accounts + 150 Enterprise
+- Identified competitive moat: cookie automation, structured reports, multi-model support
+
+### Phase 4: Technical Architecture
+- **Multi-user architecture designed:**
+  - Data model for users, workspaces, API keys, reports, usage_logs
+  - Tech stack chosen: PostgreSQL, SQLAlchemy, Supabase Auth, Stripe
+  - API endpoints designed for auth, workspaces, reports, billing
+
+- **Billing infrastructure designed:**
+  - Stripe Products and Prices structure
+  - Database schema for subscriptions and usage tracking
+  - Rate limiting middleware by tier
+  - Webhook handlers for Stripe events
+  - Pricing copy and implementation priority
+
+## Files Created
+
+1. `docs/ATLAS_PRODUCT_ARCHITECTURE.md` — Multi-user architecture, API endpoints, migration path
+2. `docs/BILLING_INFRASTRUCTURE.md` — Stripe integration, pricing tiers, rate limiting, webhooks
+
+## Current State
+
+- **Public repo (v2 branch):** Core proxy server, dashboard, security fixes, ops tooling — ready to merge
+- **Private repo (local master):** Atlas Search MCP, extension v2, cookie automation — becoming Atlas AI product
+- **PR #2:** Commented graciously, invited contributor to review v2
+- **PR #3:** Open and ready for merge (v2 → main)
+
+## Next Steps (When Ready)
+
+1. **Merge PR #3** — sync public repo with v2 improvements
+2. **Database setup** — PostgreSQL + SQLAlchemy models
+3. **Authentication** — `/auth/register`, `/auth/login`, SSO integration
+4. **Report storage** — `/reports` CRUD with sharing
+5. **Stripe integration** — Checkout flow, webhooks, customer portal
+6. **Marketing site** — Separate from open-source repo, pricing page
+
+## Key Numbers
+
+- $10M ARR = ~$833K/month
+- 2,800 Pro @ $29/mo = $81K (10%)
+- 1,000 Team @ $299/mo = $299K (36%)
+- 150 Enterprise @ $3K/mo = $450K (54%)
+- Conversion from free to paid: 5% (industry standard for dev tools)
+- MCP ecosystem: 10K+ servers, 97M downloads/month, no monetization yet (opportunity)
+
+## The Fork in the Road
+
+You now have two codebases:
+- **`anombyte93/perplexity-api-simple`** (public) — Open source proxy, builds community
+- **`atlas-ai-private`** (create this) — Commercial product with Atlas Search, billing, multi-user
+
+The open-source repo drives adoption. The private repo captures enterprise value.
+
+---
+
+Generated: 2026-02-17
+Session: PR #2 Analysis → Open Core Strategy → Atlas AI Product Design