llmquant-data — Claude Code skill: SEC, 13F, macro
A router skill for LLMQuant Data requests: it reads the task and opens the right workflow for SEC filings, 13F holders or macro snapshots.
npx skills add LLMQuant/skills --skill llmquant-data -a claude-codeWhat llmquant-data does
The skill acts as a router. It first pins down the entity, ticker, macro indicator, period and the deliverable you want, then picks the closest workflow and opens only that one. Four workflows sit behind it: business, risk and MD&A evidence from a 10-K; top 13F managers holding a ticker plus crowding signals; a compact U.S. macro regime snapshot; and a market-facing macro brief built from macro, market and research inputs.
External facts come from LLMQuant Data. That covers reading SEC filings by section, querying 13F holder lists and ownership concentration, pulling macro indicator snapshots and histories with release dates, and fetching market prices or crypto snapshots. When LLMQuant Data or a compatible data MCP is not available, the skill asks you for the data or names the missing inputs, then continues only on retrieved or user-provided evidence and labels inference separately.
That shows up in the answer: returned dates, filing periods and coverage notices are reported, so you can see what the output rests on. Worth installing if you already run LLMQuant Data or plan to wire up your own source.
What it helps with
- pull risk factors and MD&A out of a company's 10-K
- list top 13F managers holding a ticker
- build a U.S. macro regime snapshot
- draft a market-facing macro brief
- keep retrieved facts separate from inference
Example prompts for the agent
- review the risk factors and MD&A from the latest 10-K for NVDA
- which 13F managers hold this ticker, and is there crowding
- build a compact U.S. macro snapshot for the last quarter
How to install llmquant-data
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-data -a claude-code2Manually: put the files in place
~/.claude/skills/llmquant-data/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/llmquant-data/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-data -a codex2Manually: put the files in place
.agents/skills/llmquant-data/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-data -a cursor2Manually: put the files in place
.cursor/skills/llmquant-data/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install LLMQuant/skills llmquant-data2Manually: put the files in place
.github/skills/llmquant-data/SKILL.md1Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-data -a gemini-cli2Platform CLI
gemini skills install https://github.com/LLMQuant/skills --path skills/llmquant-dataManually: put the files in place
.devin/skills/llmquant-data/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add LLMQuant/skills --skill llmquant-data -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-data
How do I install llmquant-data in Claude?
Run npx skills add LLMQuant/skills --skill llmquant-data -a claude-code. The skill then lives at ~/.claude/skills/llmquant-data/SKILL.md.
What does llmquant-data do?
It extracts the entity, ticker, indicator and period from your request, then opens one of four workflows: 10-K review, 13F holders, U.S. macro snapshot or macro brief.
What does the skill need to work?
Access to LLMQuant Data or a compatible data MCP. Without either, it asks you for the data or names the missing inputs and proceeds only on confirmed evidence.
Source text
SKILL.md · MIT
--- name: llmquant-data description: Router skill for LLMQuant Data primitive workflows. Use when the user needs SEC filings, 13F holders, macro snapshots, or source-grounded macro briefs. input_data_source: LLMQuant Data category: data --- # LLMQuant Data This category routes data-primitive research tasks to focused workflows. Use it when the user needs a direct LLMQuant Data-backed answer before any higher-level strategy or portfolio overlay. ## Routing Rules 1. Identify the user's entity, ticker, macro indicator, period, and requested deliverable. 2. Select the closest workflow below. 3. Open only that workflow and any explicitly referenced local resources. 4. Use LLMQuant Data as the source of external facts. 5. Report returned dates, filing periods, coverage notices, and missing inputs. ## Workflow Index | User intent | Workflow | |---|---| | Review business, risk, and MD&A evidence from a company's 10-K. | [`workflows/10k-risk-review.md`](workflows/10k-risk-review.md) | | Identify top 13F managers holding a ticker and crowding signals. | [`workflows/ticker-smart-money-holders.md`](workflows/ticker-smart-money-holders.md) | | Build a compact U.S. macro regime snapshot. | [`workflows/us-macro-snapshot.md`](workflows/us-macro-snapshot.md) | | Compose a market-facing macro brief from macro, market, and research inputs. | [`workflows/macro-brief.md`](workflows/macro-brief.md) | ## LLMQuant Data Contract Prefer LLMQuant Data when available. The workflows may need these data capabilities: - Read SEC filings and specific filing sections such as business, risk factors, and MD&A. - Query 13F holder lists, manager holdings, and ownership concentration for a ticker. - Retrieve macro indicator snapshots, histories, release dates, and metadata. - Retrieve market prices, crypto snapshots, research knowledge, and paper/wiki context when relevant. Fallback: - If LLMQuant Data or a compatible data MCP is unavailable, ask for user-provided data or name the missing inputs. - Continue only with retrieved or user-provided evidence and label inference separately.
© LLMQuant · MIT · LLMQuant/skills
