MCP server that checks options trades for hidden catalysts (earnings, FDA, legal, SEC) before you sell premium
Project description
TickerRisk MCP Server
Check an options trade for hidden catalysts before you sell premium — from inside Claude, ChatGPT, or Cursor.
An MCP server that lets an AI assistant answer questions like "is it safe to sell a 30-day put on HPE?" with real event data instead of a guess.
The problem it solves: a fat option premium is usually the market pricing in a known upcoming event — an earnings report, an FDA decision, a court date — not free money. Most option screeners rank by yield and never check the calendar. This one scores every candidate for catalysts that land inside the expiry window, and filters out the traps.
You: Is it safe to sell a 4-week put on INTC?
Claude: [calls scan_ticker]
INTC scores 100/100 (HIGH) over a 4-week window. Earnings land in 4 days,
inside your expiry. The premium is pricing that gap — 279% annualized on
the $88 put is compensation for event risk, not an edge.
Why this is different
Risk is horizon-dependent, and that is the whole point. The same stock:
| Ticker | 1-week window | 4-week window | Why it changes |
|---|---|---|---|
| AAPL | 51 (MEDIUM) | 97 (HIGH) | Earnings sit in week 3 |
| KO | 22 (LOW) | 62 (MEDIUM) | Earnings enter the window |
Sell a weekly and you are fine. Sell a monthly on the same ticker and you have sold straight through an earnings report. A screener that shows one number cannot tell you that.
Install
Claude Desktop
Add to your claude_desktop_config.json:
macOS/Linux — ~/Library/Application Support/Claude/claude_desktop_config.json
Windows — %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"tickerrisk": {
"command": "uvx",
"args": ["tickerrisk-mcp"]
}
}
}
Restart Claude Desktop. You should see the TickerRisk tools in the tool menu.
Claude Code
claude mcp add tickerrisk -- uvx tickerrisk-mcp
Cursor
In ~/.cursor/mcp.json:
{
"mcpServers": {
"tickerrisk": { "command": "uvx", "args": ["tickerrisk-mcp"] }
}
}
From source
git clone https://github.com/Tickerrisk/tickerrisk-mcp
cd tickerrisk-mcp
pip install -e .
tickerrisk-mcp
Tools
| Tool | What it answers |
|---|---|
scan_ticker |
"How risky is selling an option on X over the next N weeks?" |
find_wheel_candidates |
"What cash-secured puts can I sell this week without an earnings trap?" |
find_covered_calls |
"What calls can I sell against shares I already own?" |
compare_tickers |
"Which of these stocks is safest to sell premium on right now?" |
scan_ticker(ticker, expiry_weeks=4)
Returns a 0–100 catalyst-risk score (higher = riskier) with the events driving it: earnings date and whether it falls in the window, FDA/clinical milestones, legal filings, SEC events, implied volatility, IV Rank, and the expected move.
Bands: ≥70 HIGH · 45–69 MEDIUM · <45 LOW
find_wheel_candidates(week, risk, max_risk, min_put_oi, sector, limit)
Scans the S&P 500 for cash-secured puts and returns only names whose catalyst score over
the option's own expiry window is under max_risk. Flags any candidate whose earnings land
before expiry.
find_covered_calls(week, risk, max_risk, min_call_oi, sector, limit)
Same gating for the call side. Income is computed from time value only, so in-the-money strikes do not show inflated yields.
compare_tickers(tickers, expiry_weeks=4)
Side-by-side catalyst risk for up to 25 symbols on one horizon.
Access and authentication
No signup needed to start. Access follows tickerrisk.com's normal model:
- First 24 hours — full access, no account, keyed to your IP
- After that — a free account at tickerrisk.com adds 14 days
- The strategy scanners (
find_wheel_candidates,find_covered_calls,compare_tickers) read cached data and stay available
To authenticate an existing account, set a token:
{
"mcpServers": {
"tickerrisk": {
"command": "uvx",
"args": ["tickerrisk-mcp"],
"env": { "TICKERRISK_TOKEN": "your-jwt-here" }
}
}
}
Environment variables
| Variable | Default | Purpose |
|---|---|---|
TICKERRISK_TOKEN |
(none) | JWT for an existing account |
TICKERRISK_BASE_URL |
https://tickerrisk.com |
Override the API host |
TICKERRISK_TIMEOUT |
45 |
Request timeout in seconds |
Data and limitations
Being straight about what this is and is not:
- Catalyst data — earnings dates, SEC filings, court records (CourtListener), ClinicalTrials.gov, and news. Public sources, so incomplete or delayed entries happen.
- Option premiums are indicative, not live NBBO. Quotes are roughly 15 minutes delayed and estimated where no bid exists (marked in the output). Always confirm in your broker before trading.
- Outside US market hours (9:30–16:00 ET) quotes go stale and candidate lists thin out. This is upstream data reality, not a bug.
- The score is not a prediction. It measures scheduled event exposure and volatility. A LOW score does not mean a stock cannot drop; it means no known catalyst was found in that window.
Not financial advice. For research only.
How it compares
Option Samurai, Barchart, and Market Chameleon are mature screeners with far more filters and real-time data feeds — if you want depth of screening, use those. What none of them do is check whether a scheduled catalyst falls inside your specific expiry window and gate results on it. That single check is what this tool exists for, and it is the reason a 100/100 INTC shows up as a warning rather than the top yield pick.
Links
- Website — tickerrisk.com
- Wheel scanner — tickerrisk.com/wheel
- Covered calls — tickerrisk.com/covered-calls
License
MIT
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file tickerrisk_mcp-0.1.0.tar.gz.
File metadata
- Download URL: tickerrisk_mcp-0.1.0.tar.gz
- Upload date:
- Size: 10.3 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
ba0005e4c87b2d57f848bcc7b8be1467955fab13116a32fd1ba65e396c6e382b
|
|
| MD5 |
3289bf0548824932d2b55862207ea946
|
|
| BLAKE2b-256 |
4669c6e8a5d5aa8a0c7e9d4d33d1c7a774393e09b83539a6df177fab1fdc1eac
|
File details
Details for the file tickerrisk_mcp-0.1.0-py3-none-any.whl.
File metadata
- Download URL: tickerrisk_mcp-0.1.0-py3-none-any.whl
- Upload date:
- Size: 12.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.13.3
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
717febbe73ef27f45137d29bb99d8e2eda28138c259f467e21f3b449be01f939
|
|
| MD5 |
f1676188bbfed602eac0ad329066377e
|
|
| BLAKE2b-256 |
ab6606f31b6fb2cc8c4b630388a41bee7f844a3e61d636578a2a017223335bbc
|