sec-edgar
starsforks
How to install sec-edgar?
kansoku-trade/kansoku/sec-edgar
$npx -y skills add kansoku-trade/kansoku --skill sec-edgarUS SEC EDGAR filings — list 10-K/10-Q/8-K/Form 4/S-1, fetch filing text, parse insider Form 4 transactions.
SKILL.md
| 1 | # sec-edgar |
| 2 | |
| 3 | > Response language: match user input. |
| 4 | |
| 5 | ## When to use |
| 6 | |
| 7 | Trigger phrases: |
| 8 | |
| 9 | - 8-K / 10-K / 10-Q / Form 4 / S-1 / proxy / DEF 14A |
| 10 | - 美股公告 / SEC filing / EDGAR |
| 11 | - insider trading / 内部人交易 / 高管交易 |
| 12 | - 财报原文 / 招股书 / 风险因素 / MD&A |
| 13 | |
| 14 | **Not for**: 13F holdings analysis (this skill lists 13F filings but does not |
| 15 | parse them into holdings — deferred to a future skill). |
| 16 | |
| 17 | ## Workflow |
| 18 | |
| 19 | 1. **List filings**: `filings.py <TICKER>` (optionally `--type 8-K`). |
| 20 | 2. **Read filing text**: pass `primary_doc_url` from the list output to |
| 21 | `filing_text.py`. Use `--max-chars` to cap; `--section item1a` for risk |
| 22 | factors, `--section mda` (Item 7) for MD&A. |
| 23 | 3. **Insider trades**: `insider.py <TICKER>` parses Form 4 XML for the |
| 24 | past --days window. |
| 25 | |
| 26 | Environment auto-loaded; `SEC_USER_AGENT` is mandatory. |
| 27 | |
| 28 | ## CLI examples |
| 29 | |
| 30 | ```bash |
| 31 | # Latest 5 NVDA 8-Ks |
| 32 | python3 .claude/skills/sec-edgar/scripts/filings.py NVDA --type 8-K --limit 5 |
| 33 | |
| 34 | # All recent NVDA filings |
| 35 | python3 .claude/skills/sec-edgar/scripts/filings.py NVDA --limit 20 |
| 36 | |
| 37 | # Read the most recent 8-K (URL from filings.py) |
| 38 | python3 .claude/skills/sec-edgar/scripts/filing_text.py \ |
| 39 | "https://www.sec.gov/Archives/edgar/data/1045810/000104581026000051/nvda-20260520.htm" \ |
| 40 | --max-chars 20000 |
| 41 | |
| 42 | # Pull Item 1A (Risk Factors) from a 10-K |
| 43 | python3 .claude/skills/sec-edgar/scripts/filing_text.py "<10-K URL>" --section item1a |
| 44 | |
| 45 | # Save full text to disk, get metadata only |
| 46 | python3 .claude/skills/sec-edgar/scripts/filing_text.py "<URL>" --save-raw /tmp/nvda-10k.txt |
| 47 | |
| 48 | # Insider transactions, past 90 days |
| 49 | python3 .claude/skills/sec-edgar/scripts/insider.py NVDA --days 90 |
| 50 | |
| 51 | # Insider + amendments |
| 52 | python3 .claude/skills/sec-edgar/scripts/insider.py NVDA --include-amendments |
| 53 | ``` |
| 54 | |
| 55 | ## Section keys (10-K) |
| 56 | |
| 57 | | Key | 10-K section | |
| 58 | | -------------------- | ----------------------------------------------- | |
| 59 | | `item1` / `business` | Item 1. Business | |
| 60 | | `item1a` / `risk` | Item 1A. Risk Factors | |
| 61 | | `item7` / `mda` | Item 7. MD&A | |
| 62 | | `item7a` | Item 7A. Quantitative & Qualitative Disclosures | |
| 63 | | `item8` | Item 8. Financial Statements | |
| 64 | |
| 65 | `meta.confidence` returned: `high` (clean heading match), `medium` (short |
| 66 | slice, may be incomplete), `low` (heuristic fallback — full text returned |
| 67 | with warning). |
| 68 | |
| 69 | ## Output shapes |
| 70 | |
| 71 | `filings.py`: |
| 72 | |
| 73 | ```json |
| 74 | { |
| 75 | "data": [ |
| 76 | { |
| 77 | "accession": "0001045810-26-000051", |
| 78 | "cik": "0001045810", |
| 79 | "form": "8-K", |
| 80 | "filed_date": "2026-05-20", |
| 81 | "primary_doc_url": "https://www.sec.gov/Archives/edgar/data/1045810/...htm", |
| 82 | "primary_doc_name": "nvda-20260520.htm", |
| 83 | "description": "8-K", |
| 84 | "size": 637530, |
| 85 | "is_xbrl": true |
| 86 | } |
| 87 | ], |
| 88 | "meta": { "cik": "0001045810", "name": "NVIDIA CORP", "count_returned": 1 }, |
| 89 | "ok": true |
| 90 | } |
| 91 | ``` |
| 92 | |
| 93 | `insider.py`: |
| 94 | |
| 95 | ```json |
| 96 | { |
| 97 | "data": [ |
| 98 | { |
| 99 | "accession": "...", |
| 100 | "form": "4", |
| 101 | "filed_date": "2026-05-15", |
| 102 | "reporter": "JEN-HSUN HUANG", |
| 103 | "roles": ["officer:CEO", "director"], |
| 104 | "txn_date": "2026-05-13", |
| 105 | "security_title": "Common Stock", |
| 106 | "code": "S", |
| 107 | "shares": 240000, |
| 108 | "price": 412.5, |
| 109 | "acquire_or_dispose": "D", |
| 110 | "post_holdings": 78000000, |
| 111 | "ownership_kind": "D", |
| 112 | "derivative": false, |
| 113 | "footnote_ids": ["F1"], |
| 114 | "footnotes_text": ["Sale pursuant to 10b5-1 plan adopted ..."] |
| 115 | } |
| 116 | ], |
| 117 | "meta": { "cik": "...", "name": "...", "filings_scanned": 12, "txns_parsed": 24 }, |
| 118 | "ok": true |
| 119 | } |
| 120 | ``` |
| 121 | |
| 122 | ## Error handling |
| 123 | |
| 124 | | Exit code | Meaning | LLM action | |
| 125 | | --------- | ---------------------------------------------- | ---------------------------------------------------------------------- | |
| 126 | | 0 | Success | Parse and narrate. | |
| 127 | | 2 | Missing `SEC_USER_AGENT` | Set in `.env` at project root. | |
| 128 | | 3 | HTTP 4xx, ticker not found, or unparseable XML | Read `hint`. Per-filing parse errors collected in `meta.parse_errors`. | |
| 129 | | 4 | Network | Suggest retry. | |
| 130 | |
| 131 | ## Known limitations |
| 132 | |
| 133 | - `insider.py` only scans the `recent` window of submissions (typically last |
| 134 | ~1000 filings); deep history requires `submissions/CIK*-N.json` paging, |
| 135 | not implemented. |
| 136 | - `filing_text.py` uses an HTML-only extractor — XBRL inline tables are |
| 137 | stripped to text, not parsed into structured rows. |
| 138 | - Section slicing is regex-heuristic; trust `meta.confidence`. |
| 139 | - 10 req/s throttle applied globally acr |
People who liked this also viewed
6 picksType
- azure-costAzure cost management: query costs, forecast spending, optimize to reduce waste. WHEN: \"Azure costs\", \"Azure bill\", \"cost breakdown\", \"how much am I spending\", \"forecast spending\", \"optimizSkillsJul 2026
- pricingWhen the user wants help with pricing decisions, packaging, or monetization strategy. Also use when the user mentions 'pricing,' 'pricing tiers,' 'freemium,' 'free trial,' 'packaging,' 'price increaseSkillsJul 2026
- stripe-integrationMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- startup-financial-modelingMulti-harness agentic plugin marketplace for Claude Code, Codex CLI, Cursor, OpenCode, GitHub Copilot, and Gemini CLISkillsJul 2026
- tushare-finance专为 Claude Code 设计的 Tushare Pro 金融数据技能包，支持 220+ 个接口，用自然语言获取 A 股行情、财务数据、宏观经济等金融数据。SkillsJul 2026
- azure-pricingFetches real-time Azure retail pricing using the Azure Retail Prices API (prices.azure.com) and estimates Copilot Studio agent credit consumption. Use when the user asks about the cost of any Azure seSkillsJul 2026
