Guide · Agent skill
How to use the SEC EDGAR Filing Analysis agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The edgar-sec-filings skill is a set of instructions that teaches an AI agent how to read US public company filings from the SEC's EDGAR database. It covers annual reports (10-K), quarterly reports (10-Q), current events (8-K), proxy statements (DEF 14A) and insider Form 4 transactions. After following this guide you will know how to install the skill, ask for filing analysis in plain language, and what the output can and cannot be trusted for.
SEC EDGAR Filing AnalysisHKUDS/Vibe-Trading · MIT Download the skill Quick start
- 1Download the ZIP. Download the edgar-sec-filings skill ZIP from this page and unzip it. It contains SKILL.md.
- 2Place the skill folder. Move the folder into ~/.claude/skills/ for personal use or into a project's .claude/skills/ folder for one project.
- 3Or upload in the Claude apps. In the Claude apps, open Settings, go to Capabilities, then Skills, and upload the skill.
- 4Check requirements. Make sure Python with yfinance is available, and that the agent runtime has network access to sec.gov, efts.sec.gov and Yahoo Finance and a read_url tool.
- 5Ask a filing question. Name the company or ticker, the filing type and the period, and say what you want to know.
- 6Verify against EDGAR. Check any figure or quote against the original filing on sec.gov before you rely on it.
What the skill makes the agent do
Once installed, the agent follows a structured framework for interpreting SEC filings. It can pull out key financials such as revenue growth, gross margin trend, cash flow, free cash flow and debt position. It can also summarise management's discussion and analysis (MD&A) and compare risk factors against a prior filing to spot newly added risks.
The skill also flags notable events such as executive departures, restatements and insider buying or selling, and explains what each filing type is generally used to signal. It is an analytical framework, not a source of verified investment advice.
- 10-K: full-year financials, risk factors, MD&A, segment data
- 10-Q: quarterly financials and interim MD&A
- 8-K: material events such as M&A, CEO change, restatement, guidance
- DEF 14A: executive compensation, board composition, shareholder proposals
- Form 4: insider buys and sells, filed within 2 days
Install the skill (ZIP download or Claude apps)
Download the skill ZIP from this page and unzip it. The skill is a folder containing a SKILL.md file. For Claude Code, place the unzipped folder in ~/.claude/skills/ to make it available everywhere, or in a project's .claude/skills/ folder to limit it to that project.
If you use the Claude apps instead, open Settings, go to Capabilities, then Skills, and upload the skill there.
- Option A (personal): unzip into ~/.claude/skills/
- Option B (single project): unzip into the project's .claude/skills/
- Option C (Claude apps): Settings → Capabilities → Skills, then upload
~/.claude/skills/What the skill needs to run
The skill expects Python with the yfinance library, which it uses for structured financial statements and insider transaction data. It also needs network access to sec.gov, efts.sec.gov and Yahoo Finance data through yfinance, and a read_url tool in the agent runtime for fetching EDGAR pages.
Without network access the agent cannot retrieve the filings, so it can only discuss the framework in general terms.
- Python with the yfinance library installed
- Network access to sec.gov, efts.sec.gov and Yahoo Finance (via yfinance)
- A read_url tool available in the agent runtime
How to trigger the skill with natural requests
You do not need special syntax. Name the company or ticker, the filing type and what you want to know. The agent should pick up the skill from the request when it matches the skill's description.
Be specific about the period so the agent knows which filing to compare against. Asking for a change from the prior year works better than asking for a general opinion.
- "Summarise Apple's latest 10-K: revenue trend, margins and cash flow."
- "Compare the risk factors in Company X's most recent 10-K with the previous year."
- "Check the last 90 days of insider Form 4 activity for AAPL and say whether there was cluster buying."
- "Did Company X file an 8-K about a CEO or CFO departure this quarter?"
Safety notes: scripts and network access
This skill runs scripts and uses the network. The source shows Python snippets that call yfinance and EDGAR URLs. Review what the agent runs before you approve it, especially if you use it in a project with sensitive files.
The signal scoring in the skill is simplified. Its tone analysis counts words from fixed lists, and its insider and 8-K rules are fixed heuristics, not validated predictive models. Do not treat these outputs as trading decisions or as a substitute for reading the original document.
- Runs scripts and makes network requests: check before approving
- Signals are simple heuristics, not validated models
- Non-US companies are not covered
- For legal or compliance decisions, have a qualified professional review the original filing
Tips for better results
Ask for the original filing link or the specific section you want quoted, so you can check the agent's reading against the source. Compare filings of the same type over time, since the skill's most useful outputs are changes rather than single-period levels.
If the agent gives a signal, ask it to show the numbers behind it. Word counts and fixed thresholds are easier to judge when you can see them.
- Ask for the exact section (for example, the MD&A) and check the quote against EDGAR
- Compare the same filing type across periods
- Ask for the numbers behind any signal
What people use it for
- First-pass read of a company's latest 10-K or 10-Q for revenue, margins and cash flow
- Spotting new or intensified risk factors by comparing this year's 10-K with last year's
- Reviewing recent 8-K events such as executive departures or restatements
- Scanning recent Form 4 insider buying and selling for clusters
- Learning how to interpret financial statements and SEC disclosures
Frequently asked questions
How do I install the edgar-sec-filings skill?
Download the skill ZIP from this page and unzip it. Put the folder in ~/.claude/skills/ for personal use or in a project's .claude/skills/ folder. In the Claude apps, you can instead upload it under Settings, Capabilities, Skills.
What do I need to run it?
Python with the yfinance library, network access to sec.gov, efts.sec.gov and Yahoo Finance via yfinance, and a read_url tool in the agent runtime. Without network access the agent cannot fetch the filings.
Is the edgar-sec-filings skill free?
The skill is listed under the MIT licence. Check the licence file in the source repository for the full terms.
Does it work for non-US companies?
No. The skill covers US public company filings only, so it is not suitable for companies that do not file with the SEC's EDGAR system.
Can I use it to decide whether to buy or sell a stock?
It is not designed for that. The signals are simplified heuristics such as word counts and fixed rules, not validated predictive models. Use it for a first-pass read and check the original filing before making any decision.
What is the difference between a 10-K and a 10-Q?
A 10-K is the annual report with full-year financials, risk factors and MD&A. A 10-Q is the quarterly report with interim financials and updates, which the skill uses to confirm trends or spot inflections.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you want a structured first-pass read of US company filings and you are comfortable checking the source yourself. Pick something else if you need trading signals you can rely on, coverage of non-US companies, or a tool for legal or compliance decisions where a qualified professional must review the original document.
Download the skill