What is LangAlpha AGENTS.md?

Project guidance for LangAlpha, the AI-agent service in a financial research platform. It describes a research-focused worker agent, a lightweight lookup assistant, and the backend and frontend structure.

Use it when working on the Python backend, web frontend, agent behavior, financial research tools, sandboxes, or local development commands.

It gives coding agents the project's architecture, shared instructions, and common commands for running or changing the full stack.

LangAlpha AGENTS.md — Instructions file for Codex and OpenCode by ginlix-ai: Project guidance for LangAlpha, the AI-agent service in a…

About the project

LangAlpha is an agent workspace for researching financial markets and supporting investment decisions through persistent research, news analysis, and parallel subagents. It is for investors who want to develop and update trading theses over time, including generating long-short pair-trade ideas. The catalogue entries provide the skills, instructions, MCP servers, and plugin that make up its agent workflow.

ginlix-ai/LangAlpha · 1,720 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/ginlix-ai/langalpha/agents-md

Clone the repo

git clone --depth 1 https://github.com/ginlix-ai/LangAlpha

Made for: Codex, OpenCode.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/ginlix-ai/langalpha/agents-md.svg)](https://agentmods.dev/instructions/ginlix-ai/langalpha/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/ginlix-ai/langalpha/agents-md"><img src="https://agentmods.dev/badge/instructions/ginlix-ai/langalpha/agents-md.svg" alt="Measured on agentmods" height="20"></a>

Per session 2,596 This file is loaded in full into every session.

When invoked 2,596 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.02596	$0.02596
Opus 5	$0.01298	$0.01298
Sonnet 5	$0.00519	$0.00519
Haiku 4.5	$0.00260	$0.00260

Measured 2d ago against content hash aa2714b7f8f2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

LangAlpha AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

AGENTS.md · 93 lines

How it starts

The opening of the file, as written. The whole thing — 93 lines — stays where its author put it; the contents beside it link to each section on GitHub.

langalpha

Core AI agent service of the Ginlix financial research platform. Two agents:

PTC agent (default) — the worker: does the R&D and produces deliverables. Wired with a full Daytona sandbox and the complete toolset (code execution, MCP financial-data tools, charts, subagent orchestration). PTC = Programmatic Tool Calling (see PTC pattern).
Flash agent — a fast, lightweight assistant: quick lookups, and coordinating between the workspace and the PTC worker. No sandbox; external tools only.

Single source of truth for AI coding agents. CLAUDE.md imports this via @AGENTS.md; Codex/Cursor/Copilot read it directly. Edit here, not there.

Common Commands

make up                               # full stack (postgres, redis, backend, frontend); auto-detects sandbox
make config                           # interactive setup wizard (LLM, data, sandbox, search)
make help                             # all targets (deploy, prod-up, test-sandbox, …)

uv run python server.py --reload      # backend only, port 8000 (needs DB + Redis)
cd web && pnpm dev                    # frontend dev server, port 5173

make lint                             # Ruff (backend) + ESLint (frontend)

# Backend tests — default run is unit only (integration/slow/regression deselected in pyproject)
uv run pytest tests/unit/ -v --tb=short
uv run pytest -m integration          # hits real APIs — needs DB + Redis + API keys
uv run pytest -m regression           # locks live market-data behavior — needs a running server + live providers

cd web && pnpm test                   # Vitest;  pnpm test:e2e = Playwright;  pnpm typecheck = tsc --noEmit

Architecture

Backend (`src/`)

Directory	Purpose
`src/server/`	FastAPI app, routers (`app/`), handlers, models, services
`src/ptc_agent/`	Core agent library — factory, middleware stack, subagents, prompts, sandbox/MCP integration
`src/tools/`	LangChain tools: web search/fetch, market data, SEC filings, crawl
`src/llms/`	LLM wrappers, token counting, pricing, model manifest (`manifest/models.json`)
`src/data_client/`	Financial data protocol abstraction
`src/utils/`	Redis cache, shared utilities
`libs/ptc-cli/`	Standalone interactive CLI for the PTC agent (pkg `langalpha-cli`, cmd `ptc-agent`)

Read the full file on GitHub · 93 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago Changed · +1 lines · +99 tokens per session aa2714b7f8f2
5d ago First seen · 92 lines · 2,497 tokens per session scan A 4ee4cefc0ced

Subscribe to this mod's changes

LangAlpha AGENTS.md is an instructions file published in the GitHub repository ginlix-ai/LangAlpha (1,720 stars, last pushed today), licensed Apache-2.0. It adds 2,596 tokens to every session, about $0.0130 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Backend and APIs Frontend LLM LangChain Planning agent investment langraph mcp

LangAlpha AGENTS.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

langalpha

Common Commands

Architecture

Backend (`src/`)

What this file has done since we first saw it

Other instructions, from other repositories

zero2Agent AGENTS.md

InvestSkill copilot-instructions.md

InvestSkill GEMINI.md

flock AGENTS.md

flock CLAUDE.md

InvestSkill CLAUDE.md

LangAlpha AGENTS.md

Getting it into your agent

Wrote this? Show the measurements

What it costs to keep this loaded

Grade A, and why

Nothing flagged

How it starts

langalpha

Common Commands

Architecture

Backend (src/)

What this file has done since we first saw it

Other instructions, from other repositories

zero2Agent AGENTS.md

InvestSkill copilot-instructions.md

InvestSkill GEMINI.md

flock AGENTS.md

flock CLAUDE.md

InvestSkill CLAUDE.md

Backend (`src/`)