scraperapi-java-sdk
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill facilitates indirect prompt injection by design, as it provides methods to ingest untrusted data from external websites.
- Ingestion point: The
client.get(url).result()method inSKILL.mdfetches arbitrary content from user-provided URLs. - Boundary markers: The provided code examples do not demonstrate the use of delimiters or 'ignore' instructions to isolate scraped content from the agent's logic.
- Capability inventory: The skill utilizes
ScraperApiClientandjava.net.http.HttpClientfor network operations inSKILL.md. - Sanitization: Examples return raw HTML strings or JSON without demonstrating sanitization or filtering of the external data.
- [COMMAND_EXECUTION]: The skill provides instructions for executing Java code that performs network requests to ScraperAPI's official infrastructure (
scraperapi.com,async.scraperapi.com) to facilitate web scraping. - [EXTERNAL_DOWNLOADS]: The documentation references the
com.scraperapi:sdkMaven/Gradle artifact and the Jackson JSON library for project setup and dependency management.
Audit Metadata