The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes multiple specialized Python scripts located within the ${CLAUDE_PLUGIN_ROOT}/scripts/claude-seo directory. These tools, including moz_api.py, bing_webmaster.py, and verify_backlinks.py, are used to perform authenticated API calls and crawl web data.\n- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from websites and APIs, which constitutes a potential attack surface for indirect prompt injection.\n
Ingestion points: Untrusted content is ingested from target websites via the Verification Crawler and through anchor text and metadata retrieved from external SEO APIs.\n
Boundary markers: The skill utilizes a structured 7-section analysis framework to organize findings, which provides context-based separation of data.\n
Capability inventory: The skill has permissions to execute local scripts, perform network requests to verified API endpoints, and conduct web crawling.\n
Sanitization: A mandatory validation gate script (validate_backlink_report.py) is executed to verify data integrity and ensure that all metrics are correctly labeled with their respective sources and confidence scores before the report is presented.\n- [EXTERNAL_DOWNLOADS]: The documentation references a local installation script (./extensions/dataforseo/install.sh) for an optional premium extension. This script is used to configure integration with the DataForSEO service.

seo-backlinks