Seeking Alpha Skill

Access Seeking Alpha (SA) data using authenticated session cookies. This skill provides programmatic access to SA's API and HTML pages for stock research, article reading, quant ratings, and more.

Prerequisites

•Credentials: sa_credentials.json (in this directory) contains session cookies exported from a premium SA account.
•Python packages: requests (standard; no extra install needed).
•Cookie refresh: Session cookies expire. If requests return 401/403 or redirect to login, re-export cookies from the browser and update sa_credentials.json.

Quick Start

python

import json, requests, re

# Load cookies
SKILL_DIR = "/Users/henryzha/Desktop/alpha-1/sa/skills/seeking_alpha"
with open(f"{SKILL_DIR}/sa_credentials.json") as f:
    creds = json.load(f)

cookie_jar = {c["name"]: c["value"] for c in creds["cookies"]}

HEADERS = {
    "User-Agent": "Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/131.0.0.0 Safari/537.36",
    "Accept": "application/json",
}

BASE = "https://seekingalpha.com"

def sa_get(path, params=None):
    """Make authenticated GET request to SA API."""
    return requests.get(f"{BASE}{path}", headers=HEADERS, cookies=cookie_jar, params=params, timeout=15)

Working API Endpoints

All endpoints are GET requests to https://seekingalpha.com/api/v3/....

1. Symbol Info

code

GET /api/v3/symbols/{TICKER}

Returns: company name, exchange, equity type, followers count, etc.

Param	Example	Description
`include`	`peers`	Include related data

Response fields: companyName, exchange, currency, equityType, followersCount, tradingViewSlug, divYieldType, etc.

2. Tickers (Batch Lookup)

code

GET /api/v3/tickers?filter[slugs]=AAPL,MSFT,GOOGL

Returns ticker metadata for multiple symbols at once. Same fields as Symbol Info.

3. Analysis Articles

code

GET /api/v3/symbols/{TICKER}/analysis

Lists analyst opinion articles for a ticker.

Param	Example	Description
`page[size]`	`20`	Results per page (max ~40)
`page[number]`	`2`	Page number
`filter[since]`	`0`	Unix timestamp, articles after
`filter[until]`	`0`	Unix timestamp, articles before
`include`	`author,primaryTickers,secondaryTickers`	Related data

Response fields per article: id, title, publishOn, commentCount, isPaywalled, isLockedPro, structuredInsights, themes.

4. News

code

GET /api/v3/symbols/{TICKER}/news

Same params as Analysis. Returns news items.

5. Transcripts (Earnings Calls)

code

GET /api/v3/symbols/{TICKER}/transcripts

Same params as Analysis. Returns earnings call transcripts.

6. Press Releases

code

GET /api/v3/symbols/{TICKER}/press-releases

Same params as Analysis.

7. Related Analysis

code

GET /api/v3/symbols/{TICKER}/related-analysis

Articles related to the ticker from other authors.

8. Quant Ratings

code

GET /api/v3/symbols/{TICKER}/ratings

Returns SA's quant rating grades - this is one of the most valuable endpoints.

Response ratings object fields:

Field	Type	Description
`quantRating`	float	Overall quant rating (1-5 scale, 5=Strong Buy)
`authorsRating`	float	SA authors consensus rating
`sellSideRating`	float	Wall Street analysts consensus
`valueGrade`	int	Valuation grade (1=A+, 13=F)
`growthGrade`	int	Growth grade
`profitabilityGrade`	int	Profitability grade
`momentumGrade`	int	Momentum grade
`epsRevisionsGrade`	int	EPS revisions grade
`dividendYieldGrade`	int	Dividend yield grade
`divSafetyCategoryGrade`	int	Dividend safety grade
`divGrowthCategoryGrade`	int	Dividend growth grade
`divConsistencyCategoryGrade`	int	Dividend consistency grade

Grade mapping: 1=A+, 2=A, 3=A-, 4=B+, 5=B, 6=B-, 7=C+, 8=C, 9=C-, 10=D+, 11=D, 12=D-, 13=F

Author rating counts: authorsRatingBuyCount, authorsRatingHoldCount, authorsRatingSellCount, authorsRatingStrongBuyCount, authorsRatingStrongSellCount (also available with 30Day and 90Day suffixes).

9. Dividend History

code

GET /api/v3/symbols/{TICKER}/dividend_history

Returns full dividend payment history.

Param	Example	Description
`page[size]`	`20`	Results per page

Response fields: amount, adjusted_amount, ex_date, pay_date, record_date, declare_date, freq, type, year, split_adj_factor.

10. Article Detail (Preview)

code

GET /api/v3/articles/{ARTICLE_ID}

Returns article metadata and a content preview (~300 words of HTML). Full content requires browser rendering.

Response fields: title, publishOn, content (HTML preview), summary (bullet points array), disclosure, commentCount, likesCount, isPaywalled, isLockedPro, and more.

Important: The summary field is an array of bullet-point strings summarizing the article - very useful for quick analysis without needing the full text.

11. Market Status

code

GET /api/v3/market_open

Returns: marketOpen (bool), nextMarketOpen, nextMarketClose (timestamps).

Extracting Full Article Text (HTML Scraping)

The API only returns article previews. For full text, scrape the article HTML page:

python

def get_full_article(article_id):
    """Get full article text by scraping the HTML page."""
    h = {**HEADERS, "Accept": "text/html"}
    r = requests.get(f"{BASE}/article/{article_id}", headers=h, cookies=cookie_jar, timeout=15)

    # Extract from SSR_DATA (server-side rendered preview)
    m = re.search(r'window\.SSR_DATA\s*=\s*({.*?});\s*</script>', r.text, re.DOTALL)
    if m:
        ssr = json.loads(m.group(1))
        attrs = ssr.get("article", {}).get("response", {}).get("data", {}).get("attributes", {})
        content_html = attrs.get("content", "")
        summary = attrs.get("summary", [])
        title = attrs.get("title", "")
        # Strip HTML tags
        text = re.sub(r'<[^>]+>', '', content_html).strip()
        return {"title": title, "text": text, "summary": summary}
    return None

Note: SSR_DATA contains a preview (~300 words). Full article body is hydrated client-side by React. For complete text, you would need a headless browser (Playwright/Selenium). For most research purposes, the summary array + preview text is sufficient.

Example Workflows

Get Quant Ratings for Multiple Stocks

python

tickers = ["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA"]
for t in tickers:
    r = sa_get(f"/api/v3/symbols/{t}/ratings")
    if r.status_code == 200:
        data = r.json().get("data", [])
        if data:
            ratings = data[0]["attributes"]["ratings"]
            print(f"{t}: Quant={ratings['quantRating']:.2f}, "
                  f"Value={ratings['valueGrade']}, Growth={ratings['growthGrade']}, "
                  f"Profit={ratings['profitabilityGrade']}, Mom={ratings['momentumGrade']}")

Get Latest Analysis with Summaries

python

r = sa_get("/api/v3/symbols/AAPL/analysis", {"page[size]": "5", "include": "author"})
for art in r.json().get("data", []):
    attrs = art["attributes"]
    print(f"[{art['id']}] {attrs['title']}")
    print(f"  Published: {attrs['publishOn']}")

    # Get summary
    detail = sa_get(f"/api/v3/articles/{art['id']}")
    summary = detail.json()["data"]["attributes"].get("summary", [])
    for s in summary:
        print(f"  - {s}")
    print()

Screen Stocks by Quant Rating

python

# SA doesn't have a public screener API, but you can batch-check ratings
import time

sp500_sample = ["AAPL", "MSFT", "GOOGL", "AMZN", "NVDA", "META", "TSLA", "BRK.B", "JPM", "V"]
strong_buys = []

for t in sp500_sample:
    r = sa_get(f"/api/v3/symbols/{t}/ratings")
    if r.status_code == 200:
        data = r.json().get("data", [])
        if data:
            qr = data[0]["attributes"]["ratings"]["quantRating"]
            if qr >= 4.5:
                strong_buys.append((t, qr))
    time.sleep(0.5)  # Be polite

print("Strong Buy stocks:", strong_buys)

Rate Limiting & Best Practices

•No official rate limit documented, but be respectful: ~1-2 requests/second.
•Use time.sleep(0.5) between requests when iterating over many tickers.
•Batch where possible: Use /api/v3/tickers?filter[slugs]=AAPL,MSFT,... for basic info.
•Cache responses: Ratings and article lists don't change minute-to-minute.
•Cookie expiry: The user_remember_token cookie is the primary session cookie. If it expires (~1 year), re-login in browser and re-export.

Troubleshooting

Symptom	Cause	Fix
401/403 responses	Cookies expired	Re-export from browser, update `sa_credentials.json`
404 on known endpoint	API version changed	SA periodically updates API; check current page network requests
Empty `content` in articles	Normal behavior	API returns preview only; use `summary` field or HTML scraping
429 Too Many Requests	Rate limited	Add `time.sleep(1)` between requests
PerimeterX block (403 with captcha)	Bot detection	Ensure `_px3` and `_pxvid` cookies are included

Files in This Skill

code

seeking_alpha/
  SKILL.md              # This file
  sa_credentials.json   # Auth cookies (essential-only, cleaned from full export)
  reference/
    api_endpoints.json  # Machine-readable endpoint definitions
    grade_mapping.json  # Grade number -> letter mapping
    example_responses/  # Sample API responses for reference