Installation
$npx skills add nikhilbhansali/dataforseo-skill-claude --skill dataforseoSummary
This skill gives an agent access to DataForSEO's full suite of APIs for keyword research, SERP analysis, backlink audits, competitor analysis, rank tracking, and technical SEO audits. Invoke it when the user needs SEO data: search volume, keyword ideas, rankings, backlinks, domain authority, technical audits, or competitive intelligence.
SKILL.MD
DataForSEO API Skill
Universal interface to all DataForSEO APIs for comprehensive SEO data retrieval and analysis.
Credential Setup
Before first use, set up credentials:
import sys, os
sys.path.insert(0, os.path.expanduser('~/.agents/skills/dataforseo/scripts'))
from dataforseo_client import save_credentials, verify_credentials
# Get credentials from https://app.dataforseo.com/
login = "your_email@example.com" # API login (email)
password = "your_api_password" # API password (from dashboard)
# Verify and save
if verify_credentials(login, password):
save_credentials(login, password)
print("Credentials saved!")
Credentials stored at ~/.dataforseo_config.json. To update, run setup again.
Quick Start
import sys, os
sys.path.insert(0, os.path.expanduser('~/.agents/skills/dataforseo/scripts'))
from dataforseo_client import *
# Example: Get search volume
response = keywords_search_volume(
keywords=["seo tools", "keyword research"],
location_name="United States"
)
results = extract_results(response)
csv_path = to_csv(results, "keyword_volumes")
print(f"Results saved to: {csv_path}")
API Selection Guide
| User Request | Function to Use |
|---|---|
| Search volume, CPC, competition | keywords_search_volume() |
| Keyword ideas/suggestions | labs_keyword_ideas() or labs_related_keywords() |
| Keywords a site ranks for | labs_ranked_keywords() |
| SERP results for keyword | serp_google_organic() |
| Local/Maps rankings | serp_google_maps() |
| YouTube rankings | serp_youtube() |
| Backlink profile | backlinks_summary() |
| List of backlinks | backlinks_list() |
| Referring domains | backlinks_referring_domains() |
| Domain authority/rank | backlinks_bulk_ranks() |
| Competing domains | labs_competitors_domain() |
| Keyword gap analysis | labs_domain_intersection() |
| Link gap analysis | backlinks_domain_intersection() |
| Technical page audit | onpage_instant_pages() |
| Lighthouse scores | lighthouse_live() |
| Technology stack | domain_technologies() |
| Brand mentions | content_search() |
| Google Trends | google_trends() |
Core Workflow
- Import client: Add skill path and import functions
- Call API function: Pass required parameters
- Extract results: Use
extract_results(response) - Export to CSV: Use
to_csv(results, "filename")
import sys, os
sys.path.insert(0, os.path.expanduser('~/.agents/skills/dataforseo/scripts'))
from dataforseo_client import labs_ranked_keywords, extract_results, to_csv
response = labs_ranked_keywords(
target="competitor.com",
location_name="United States",
language_name="English",
limit=500
)
results = extract_results(response)
csv_path = to_csv(results, "ranked_keywords")
Default Parameters
Most functions use these defaults:
location_name: "United States" (override with "India", "United Kingdom", etc.)language_name: "English"limit: 100 (increase up to 1000 for more results)device: "desktop" (or "mobile" for SERP)
Common Location Names
- United States, United Kingdom, India, Germany, Australia, Canada
- For city-level: "New York,New York,United States", "London,England,United Kingdom"
Output
All results export to CSV at ~/dataforseo_outputs/. Files auto-named with timestamp if not specified.
Reference Files
- API Reference:
references/api_reference.md- Complete endpoint documentation - Use Cases:
references/use_cases.md- Ready-to-use code recipes
Error Handling
response = some_api_function(...)
if response.get("status_code") == 20000:
results = extract_results(response)
# Process results
else:
print(f"Error: {response.get('status_message')}")
Rate Limits & Costs
- 2000 requests/minute max
- Live methods cost more than Standard
- Check usage with
get_user_data() - Response includes
costfield
Important Notes
- Async endpoints: Some APIs (merchant, app_data, business reviews) create tasks. Check task status separately.
- Limits: Increase
limitparameter for comprehensive data (default 100, max usually 1000) - Multiple keywords: Pass as list:
keywords=["kw1", "kw2", "kw3"]