llmquant-strategies — Claude Code skill: what it does, how to install
Routes hedge-fund workflows across equity, macro, quant, and event-driven playbooks. Helps select the matching strategy workflow based on mandate and risk budget.
npx skills add LLMQuant/skills --skill llmquant-strategies -a claude-codeWhat llmquant-strategies does
This skill acts as a router for portfolio managers. It identifies the strategy type, investment horizon, mandate, benchmark, and risk budget before touching any playbooks. Once parameters are clear, the agent opens only the single matching workflow file. Keeping unused playbooks closed prevents context pollution. Available playbooks include equity long/short, event-driven arbitrage, macro, quant research, and pod-style multi-strategy allocation.
The workflows integrate with LLMQuant Data for prices, filings, borrow rates, factor exposures, and macro indicators. Workflows must report data windows, as-of dates, and coverage gaps. If borrow context or factor feeds are unavailable, the agent avoids guessing numbers. It reports the missing data point and proceeds strictly on retrieved or user-provided evidence.
What it helps with
- route portfolio management prompts to the matching hedge-fund workflow
- structure fundamental paired-book construction with factor-aware hedging
- analyze merger arbitrage and special situations using event feeds
- manage pod-style capital allocation and risk budgeting across sub-strategies
- guide systematic backtesting with disciplined overfitting controls
Example prompts for the agent
- Set up a fundamental paired-book equity portfolio with factor hedging
- I need an event-driven workflow to analyze a merger arbitrage opportunity
- Run a pod-style capital allocation model with unified risk budgeting
How to install llmquant-strategies
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-strategies -a claude-code2Manually: put the files in place
~/.claude/skills/llmquant-strategies/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/llmquant-strategies/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-strategies -a codex2Manually: put the files in place
.agents/skills/llmquant-strategies/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-strategies -a cursor2Manually: put the files in place
.cursor/skills/llmquant-strategies/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install LLMQuant/skills llmquant-strategies2Manually: put the files in place
.github/skills/llmquant-strategies/SKILL.md1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-strategies -a gemini-cli2Platform CLI
gemini skills install https://github.com/LLMQuant/skills --path skills/llmquant-strategiesManually: put the files in place
.devin/skills/llmquant-strategies/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-strategies -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about llmquant-strategies
How to install llmquant-strategies in Claude Code?
Run npx skills add LLMQuant/skills --skill llmquant-strategies -a claude-code. You can also place the skill file directly into ~/.claude/skills/llmquant-strategies/SKILL.md.
What does llmquant-strategies do?
It evaluates strategy mandate, horizon, and risk constraints to load a specialized workflow. Supported setups include equity long/short, long-biased books, event-driven deals, macro trades, quant research, and multi-strategy allocation.
How does the skill handle missing factor or borrow data?
The skill reports stale notices, data windows, and coverage gaps. If factor, borrow, event, or backtest data is missing from LLMQuant Data, it names the missing input and proceeds only with user-supplied evidence.
Source text
SKILL.md · MIT
--- name: llmquant-strategies description: Router skill for LLMQuant hedge-fund and PM strategy workflows. Use when the user needs equity long/short, long-biased, event-driven, macro, quant, or multi-strategy playbooks. input_data_source: LLMQuant Data category: strategies --- # LLMQuant Strategies This category routes hedge-fund and portfolio-manager strategy playbooks. ## Routing Rules 1. Identify strategy type, universe, mandate, horizon, benchmark, and risk budget. 2. Select the closest workflow below. 3. Open only the selected workflow and local resources explicitly referenced by that workflow. 4. Use LLMQuant Data for market, macro, filings, holdings, factor, event, options, and risk inputs. 5. Report data windows, as-of dates, stale notices, and unsupported coverage. ## Workflow Index | User intent | Workflow | |---|---| | Fundamental paired-book construction and factor-aware hedging. | [`workflows/equity-long-short.md`](workflows/equity-long-short.md) | | Concentrated long-biased ownership with structural hedges. | [`workflows/long-biased.md`](workflows/long-biased.md) | | Merger arb, spin-offs, activism, restructurings, and special situations. | [`workflows/event-driven.md`](workflows/event-driven.md) | | Cross-asset macro regime trading. | [`workflows/macro.md`](workflows/macro.md) | | Systematic strategy research, backtesting, overfitting control, and execution discipline. | [`workflows/quant.md`](workflows/quant.md) | | Pod-style capital allocation and unified risk budgeting. | [`workflows/multi-strategy.md`](workflows/multi-strategy.md) | ## LLMQuant Data Contract Prefer LLMQuant Data when available. The workflows may need these data capabilities: - Retrieve prices, fundamentals, filings, macro indicators, options context, ETF holdings, factor exposures, event feeds, borrow context, and backtest inputs. - Capture strategy mandate, universe, benchmark, time horizon, risk budget, liquidity, and sizing constraints. - Report data windows, as-of dates, stale notices, unsupported coverage, and assumptions. Fallback: - If a strategy workflow needs unavailable factor, borrow, event, or backtest data, name the missing input and continue only with retrieved or user-provided evidence.
© LLMQuant · MIT · LLMQuant/skills