The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were identified. The skill's behavior is consistent with its stated purpose of SEO analysis.
[CREDENTIALS_UNSAFE]: The skill requires an API key (SCRAPERAPI_API_KEY) but correctly instructs that it should be handled via environment variables. This is the standard secure method for managing secrets in this context.
[DATA_EXFILTRATION]: The skill transmits search queries and web URLs to ScraperAPI's infrastructure to retrieve search results. This is the core functionality of the skill and is clearly disclosed in the description. There is no evidence of unauthorized or hidden data exfiltration.
[PROMPT_INJECTION]: The skill ingests untrusted data from the web (search results and scraped pages). This represents a standard surface for indirect prompt injection. However, the risk is assessed as safe because the skill lacks dangerous tools such as filesystem access, command execution, or code evaluation, limiting the potential impact of any malicious content encountered.
Ingestion points: mcp__ScraperAPIRemote__google_search, mcp__ScraperAPIRemote__google_news, mcp__ScraperAPIRemote__google_shopping, and mcp__ScraperAPIRemote__scrape tools in SKILL.md.
Boundary markers: The skill instructs the agent to separate data from analysis using specific headers like 'Raw SERP data' and 'Analysis', which helps the agent maintain context.
Capability inventory: Tools are restricted to network-based data retrieval; no local execution capabilities are present.
Sanitization: Relies on the model's inherent ability to distinguish between data content and system instructions.

scraperapi-serp-intelligence