Add claude GitHub actions 1778123143781

.claude-plugin/marketplace.json

@@ -0,0 +1,19 @@
+{
+    "name": "Abhishek-tools",
+    "owner": {
+        "name": "Abhishek",
+        "email": "abgupta@yahoo.com"
+    },
+
+    "plugins": [
+        {
+            "name": "independent-reviewer",
+            "source": "./independent-reviewer",
+            "description": "Carry out an independent review of all changes since last commit",
+            "version": "1.0.0",
+            "author": {
+                "name": "Abhishek"
+            }
+        }
+    ]
+}

.claude/agent-memory/commit-diff-reviewer/MEMORY.md

@@ -0,0 +1,3 @@
+# Agent Memory Index
+
+- [User Profile](user_profile.md) — Senior developer building FinAlly AI trading workstation; prefers terse, direct feedback

.claude/agents/change-reviewer.md

@@ -0,0 +1,6 @@
+---
+name: change-reviewer
+description: Carry out a comprehensive review of all changes since last commit
+---
+
+The sub agent reviews all changes since the last commit. Write your results to planning/SINCE-COMMIT-REVIEW2.md

.claude/agents/codex-reviewer.md

@@ -0,0 +1,10 @@
+---
+name: codex-reviewer
+description: Carry out a comprehensive review of PLAN.md when requested using codex
+---
+
+You are using a different AI Agent to carry out a review of the document: planning/PLAN.md.
+You MUST execute the following shell command to carry out the review - do not review yourself:
+`codex exec "Please review the file planning/PLAN.md and write your results to planning/REVIEW.md"`
+This will run the review process and save the results.
+Do not review yourself. 

.claude/commands/doc-review.md

@@ -0,0 +1 @@
+Review the documentation file in the planning folder called $ARGUMENTS and add questions, clarifications, or feedback to a new section at the end, along with any opportunities to simplify.

.claude/settings.json

@@ -2,6 +2,7 @@
   "enabledPlugins": {
     "frontend-design@claude-plugins-official": true,
     "context7@claude-plugins-official": true,
-    "playwright@claude-plugins-official": true
+    "playwright@claude-plugins-official": true,
+    "independent-reviewer@Abhishek-tools": true
   }
 }

.claude/skills/cerebras/SKILL.md

@@ -1,5 +1,5 @@
 ---
-name: cerebras-inference
+name: cerebras
 description: Use this to write code to call an LLM using LiteLLM and OpenRouter with the Cerebras inference provider
 ---
 

.github/workflows/claude.yml

@@ -46,5 +46,5 @@ jobs:
           # Optional: Add claude_args to customize behavior and configuration
           # See https://github.com/anthropics/claude-code-action/blob/main/docs/usage.md
           # or https://code.claude.com/docs/en/cli-reference for available options
-          # claude_args: '--allowed-tools Bash(gh pr:*)'
+          # claude_args: '--allowed-tools Bash(gh pr *)'
 

.gitignore

@@ -205,3 +205,6 @@ cython_debug/
 marimo/_static/
 marimo/_lsp/
 __marimo__/
+
+# Claude Code local settings
+.claude/settings.local.json

README.md

@@ -1,62 +1,54 @@
 # FinAlly — AI Trading Workstation
 
-A visually stunning AI-powered trading workstation that streams live market data, simulates portfolio trading, and integrates an LLM chat assistant that can analyze positions and execute trades via natural language.
+A Bloomberg-terminal-inspired trading simulator with live market data and an AI assistant that can analyze your portfolio and execute trades via natural language.
 
-Built entirely by coding agents as a capstone project for an agentic AI coding course.
-
-## Features
-
-- **Live price streaming** via SSE with green/red flash animations
-- **Simulated portfolio** — $10k virtual cash, market orders, instant fills
-- **Portfolio visualizations** — heatmap (treemap), P&L chart, positions table
-- **AI chat assistant** — analyzes holdings, suggests and auto-executes trades
-- **Watchlist management** — track tickers manually or via AI
-- **Dark terminal aesthetic** — Bloomberg-inspired, data-dense layout
+## Quick Start
 
-## Architecture
+```bash
+cp .env.example .env
+# Edit .env: add OPENROUTER_API_KEY (required), MASSIVE_API_KEY (optional)
+./scripts/start_mac.sh
+```
 
-Single Docker container serving everything on port 8000:
+Open [http://localhost:8000](http://localhost:8000).
 
-- **Frontend**: Next.js (static export) with TypeScript and Tailwind CSS
-- **Backend**: FastAPI (Python/uv) with SSE streaming
-- **Database**: SQLite with lazy initialization
-- **AI**: LiteLLM → OpenRouter (Cerebras inference) with structured outputs
-- **Market data**: Built-in GBM simulator (default) or Massive API (optional)
+## Features
 
-## Quick Start
+- **Live price streaming** via SSE — prices flash green/red on change
+- **Simulated portfolio** — $10k virtual cash, market orders, instant fill
+- **Sparklines & charts** — per-ticker mini-charts and a detailed main chart
+- **Portfolio heatmap** — treemap sized by weight, colored by P&L
+- **AI chat** — ask questions, get analysis, execute trades in plain English
 
-```bash
-# Clone and configure
-cp .env.example .env
-# Add your OPENROUTER_API_KEY to .env
+## Architecture
 
-# Run with Docker
-docker build -t finally .
-docker run -v finally-data:/app/db -p 8000:8000 --env-file .env finally
+Single Docker container, single port (8000):
 
-# Open http://localhost:8000
-```
+- **Frontend**: Next.js static export, served by FastAPI
+- **Backend**: FastAPI + Python (uv), SQLite database
+- **Market data**: GBM simulator by default; Massive/Polygon.io API if `MASSIVE_API_KEY` is set
+- **AI**: LiteLLM → OpenRouter (Cerebras), structured outputs for trade execution
 
 ## Environment Variables
 
 | Variable | Required | Description |
 |---|---|---|
-| `OPENROUTER_API_KEY` | Yes | OpenRouter API key for AI chat |
-| `MASSIVE_API_KEY` | No | Massive (Polygon.io) key for real market data; omit to use simulator |
-| `LLM_MOCK` | No | Set `true` for deterministic mock LLM responses (testing) |
+| `OPENROUTER_API_KEY` | Yes | LLM inference via OpenRouter |
+| `MASSIVE_API_KEY` | No | Real market data; simulator used if unset |
+| `LLM_MOCK` | No | Set `true` for deterministic mock responses (testing) |
 
-## Project Structure
+## Development
 
-```
-finally/
-├── frontend/    # Next.js static export
-├── backend/     # FastAPI uv project
-├── planning/    # Project documentation and agent contracts
-├── test/        # Playwright E2E tests
-├── db/          # SQLite volume mount (runtime)
-└── scripts/     # Start/stop helpers
+```bash
+# Backend tests
+cd backend && uv run pytest -v
+
+# Frontend
+cd frontend && npm install && npm run dev
 ```
 
-## License
+## Running Tests (E2E)
 
-See [LICENSE](LICENSE).
+```bash
+cd test && docker compose -f docker-compose.test.yml up
+```

independent-reviewer/.claude-plugin/plugin.json

@@ -0,0 +1,5 @@
+{
+    "name": "independent-reviewer",
+    "description": "Carry out an independent review of all changes since last commit",
+    "version": "1.0.0"
+}

independent-reviewer/hooks/hooks.json

@@ -0,0 +1,15 @@
+{
+  "hooks": {
+    "Stop": [
+      {
+        "matcher": "",
+        "hooks": [
+          {
+            "type": "command",
+            "command": "git status"
+          }
+        ]
+      }
+    ]
+  }
+}

planning/DECISIONS.md

@@ -0,0 +1,54 @@
+# Planning Decisions
+
+This file records concrete decisions made to resolve open questions and contract gaps in `planning/PLAN.md`.
+
+## 2026-04-13
+
+### LLM configuration
+
+- `OPENROUTER_API_KEY` is required only when `LLM_MOCK=false`.
+- `LLM_MOCK=true` is a supported no-key development and test mode.
+- The backend should fail fast at startup if mock mode is off and the API key is missing.
+
+Reasoning: this preserves the intended low-friction local and CI workflow while keeping production configuration errors obvious.
+
+### Docker persistence
+
+- Local development and test runs use a bind mount from repo `db/` to `/app/db`.
+- The plan no longer mixes bind mounts with a named Docker volume.
+- `db/finally.db` is intentionally visible on the host for inspection and persistence.
+
+Reasoning: one persistence model is easier for agents to implement consistently, and host-visible SQLite data is useful for a course project.
+
+### Chat API contract
+
+- `/api/chat` returns persisted `user_message` and `assistant_message` objects, including message IDs and timestamps.
+- Action results are returned only after execution, under `assistant_message.actions`.
+- Partial failures are represented per action with `status` plus `error`, while the overall response remains `200` for valid requests.
+- The response also includes post-execution `portfolio` and `watchlist` state so the frontend can reconcile immediately.
+
+Reasoning: this removes frontend/backend ambiguity around inline confirmations, persisted message identity, and partial trade failures.
+
+### Position lifecycle
+
+- Selling a position to zero keeps the row but resets `avg_cost` to `0`.
+- A later buy in the same ticker establishes a brand-new cost basis.
+
+Reasoning: unrealized P&L after re-entry should reflect only the new position, not stale historical basis.
+
+### Trade validation
+
+- Tickers are normalized to uppercase and trimmed.
+- Unsupported symbols are rejected.
+- Quantity must be finite, positive, and no more than 4 decimal places.
+- Manual and LLM-originated trades use the exact same validation rules.
+
+Reasoning: shared validation rules prevent drift between frontend behavior, direct API usage, and AI-triggered actions.
+
+### SSE protocol
+
+- `/api/stream/prices` uses named SSE events: `snapshot`, `price`, `watchlist`, and `heartbeat`.
+- The server sends an initial `snapshot` immediately on connect.
+- Watchlist add/remove operations emit a `watchlist` event on existing connections rather than requiring reconnect.
+
+Reasoning: explicit event types give frontend and backend a stable contract for initial render, live updates, and reconnect behavior.