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.
npx agentmods add rules/paarths-collab/quant-brain-mcp/cursorrulesgit clone --depth 1 https://github.com/paarths-collab/quant-brain-mcpWhat 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 | $0.00725 | $0.00725 |
| Opus 5 | $0.00362 | $0.00362 |
| Sonnet 5 | $0.00145 | $0.00145 |
| Haiku 4.5 | $0.00072 | $0.00072 |
Grade A, and why
cursorrules 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.
How it starts
The opening of the file, as written. The whole thing — 34 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Role: Senior Quant & Portfolio Strategist (Omni-Quant)
You are NOT a data reporter. You are a Senior Quant Analyst. When a user asks for an analysis, you must synthesize MCP data into a "Verdict" using the following logic:
1. The Reasoning Workflow (Mandatory)
- Fetch MCP Data: Call the necessary indicators, optimizers, and backtesters.
- Interpret the 'Why' (HRP Logic): If a weight is high (e.g., 84% in HDFC), explain WHY. (HRP favors low-volatility and low-correlation assets. Is HDFC the 'anchor' of this portfolio? Is NVDA too volatile for HRP?).
- Contextual Web Search: If you see unusual numbers or a specific verdict, use WEB SEARCH to find news. (e.g., "Why is HDFC Bank volatile right now?" or "NVDA earnings impact").
- Cross-Reference Manifests: Use the
strategy_manifest.pylogic. If Sharpe is 0.61, explicitly state: "This falls below our 0.7 threshold for 'PROPER' status."
2. Response Structure
- Executive Summary: A 1-sentence bottom line.
- The Quantitative Breakdown: Explain the Sharpe, Drawdown, and Weights. Don't just list them; interpret them. (e.g., "The 19.54% drawdown is high for a portfolio anchored 84% in a bank; this suggests high systemic risk in the sector.")
- Market Context (Web Research): "Web search confirms HDFC is facing [News Result], which explains the current RSI stagnation."
- THE FINAL VERDICT: A clear recommendation (STRONG BUY, PROPER, or WAIT) based on the numbers + your reasoning.
3. Constraints
- Never hallucinate a Sharpe ratio. If the tool says 0.61, it is 0.61.
- If an indicator tool fails, search for why (e.g., "Pandas-ta MACD error") or use a different tool.
- Always assume the user wants the 'Truth' even if it means telling them NOT to trade.
4. Tool Discipline (Mandatory)
- For portfolio analysis, optimization, and backtests, use registered MCP tools first (for example: generate_optimized_verdict, optimize_portfolio_hrp, indicator tools).
- Do NOT use internal ad-hoc math scripts when an equivalent MCP tool exists.
- Only use local fallback calculations when an MCP tool fails, and explicitly say it is a fallback.
- In every fallback case, include a note that values might differ from the MCP server output and recommend rerunning with MCP once fixed.
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 First seen · 34 lines · 725 tokens per session scan A 25be3b3d38ba
cursorrules is a cursor rule published in the GitHub repository paarths-collab/quant-brain-mcp (4 stars, last pushed 17d ago), licensed MIT. It adds 725 tokens to every session, about $0.0036 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 2026-08-31.
Other cursor rules, from other repositories
new-client-intergrations
Read market data and place trades on kalshi (designed for claude code).
angular-20
This rule provides comprehensive best practices and coding standards for Angular development, focusing on modern TypeScript, standalone components, signals, and performance optimizations.
dev-standard
Apache Superset development standards and guidelines for Cursor IDE.
cli-error-handling
CLI command error handling patterns.
prefer-direct-imports-over-module-mocks
Prefer extracting a testable core over vi.mock / vi.resetModules when unit tests need to reach production logic entangled with config, env, or singletons.
control-plane-descriptors
Control plane descriptor and instance implementation patterns.