llmquant-crypto — Claude Code skill: what it does, how to install
Routes cryptocurrency research workflows to analyze market regimes, inspect protocol tokenomics, and monitor perpetual funding rates.
npx skills add LLMQuant/skills --skill llmquant-crypto -a claude-codeWhat llmquant-crypto does
The skill acts as a router for LLMQuant crypto workflows. When given a request, it identifies the target asset, chain, venue, and time horizon before delegating execution to one of three workflows: workflows/crypto-market-regime.md, workflows/crypto-token-research.md, or workflows/crypto-perp-funding-monitor.md.
Under the hood, it expects data capabilities from LLMQuant Data. That covers spot prices, OHLCV history, open interest, perpetual funding rates, token unlock schedules, TVL, and cross-asset macro inputs. Generated outputs must report timestamps, venue coverage, observation windows, and notices for stale figures.
Hallucination is explicitly blocked by rule. The agent is forbidden from inferring live funding, liquidity, TVL, or holder metrics from memory. If exchange or on-chain inputs are missing, the skill demands naming the omitted fields and proceeding solely on verified price or user data.
What it helps with
- monitor perpetual funding rates, basis, and open interest
- compile token research memos with tokenomics and unlock schedules
- diagnose market regimes across majors using liquidity and macro context
- detect leverage crowding and liquidation risks across venues
Example prompts for the agent
- Diagnose the crypto market regime for BTC and ETH using current liquidity and leverage inputs
- Build a token research memo with tokenomics, unlock schedule, and TVL risks
- Check perpetual funding rates, basis, and open interest crowding on major exchanges
How to install llmquant-crypto
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-crypto -a claude-code2Manually: put the files in place
~/.claude/skills/llmquant-crypto/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/llmquant-crypto/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-crypto -a codex2Manually: put the files in place
.agents/skills/llmquant-crypto/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-crypto -a cursor2Manually: put the files in place
.cursor/skills/llmquant-crypto/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install LLMQuant/skills llmquant-crypto2Manually: put the files in place
.github/skills/llmquant-crypto/SKILL.md1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-crypto -a gemini-cli2Platform CLI
gemini skills install https://github.com/LLMQuant/skills --path skills/llmquant-cryptoManually: put the files in place
.devin/skills/llmquant-crypto/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-crypto -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-crypto
How to install llmquant-crypto in Claude Code?
Run npx skills add LLMQuant/skills --skill llmquant-crypto -a claude-code in your terminal. The file is saved to ~/.claude/skills/llmquant-crypto/SKILL.md.
What does llmquant-crypto route?
It routes requests into dedicated workflows for market regime diagnosis, token due diligence, or perpetual funding and basis tracking.
How does the skill handle missing on-chain or venue data?
It prohibits guessing live metrics from memory. The agent explicitly flags missing inputs and continues analysis using only confirmed prices, macro data, or evidence provided by the user.
Source text
SKILL.md · MIT
--- name: llmquant-crypto description: Router skill for LLMQuant crypto workflows. Use when the user needs crypto market regime analysis, token research, perpetual funding, basis, leverage, liquidity, or cross-asset crypto context. input_data_source: LLMQuant Data category: crypto --- # LLMQuant Crypto This category routes crypto research and trading-context workflows. It covers market regime, token-level diligence, and perpetual funding or basis monitoring. ## Routing Rules 1. Identify the asset, chain, venue, horizon, benchmark, and requested decision. 2. Select the closest workflow below. 3. Open only that workflow and any referenced local resources. 4. Use LLMQuant Data for crypto prices, liquidity, funding, open interest, on-chain context, macro, ETF, and risk inputs. 5. Report timestamps, venue coverage, observation windows, stale notices, and unavailable future inputs. ## Workflow Index | User intent | Workflow | |---|---| | Diagnose the crypto market regime across BTC, ETH, majors, liquidity, leverage, and macro. | [`workflows/crypto-market-regime.md`](workflows/crypto-market-regime.md) | | Build a token or protocol research memo with tokenomics, usage, valuation, and risk evidence. | [`workflows/crypto-token-research.md`](workflows/crypto-token-research.md) | | Monitor perpetual funding, basis, open interest, and leverage crowding. | [`workflows/crypto-perp-funding-monitor.md`](workflows/crypto-perp-funding-monitor.md) | ## LLMQuant Data Contract Prefer LLMQuant Data when available. The workflows may need these data capabilities: - Retrieve crypto spot prices, OHLCV history, realized volatility, drawdowns, correlations, and liquidity. - Retrieve perpetual funding, basis, open interest, liquidations, exchange flows, and venue-level timestamps. - Retrieve token supply, unlock schedules, protocol usage, revenue, TVL, holder concentration, governance, and security-risk context. - Retrieve macro, rates, liquidity, ETF, and equity-market inputs that affect crypto risk appetite. Fallback: - If on-chain, funding, or venue-level data is unavailable, name the missing input and continue only with available price, macro, or user-provided evidence. - Do not infer live funding, liquidity, TVL, or holder behavior from memory.
© LLMQuant · MIT · LLMQuant/skills
