apify-competitor-intelligence
Pass
Audited by Gen Agent Trust Hub on Mar 20, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes
nodeto run a local utility script (run_actor.js) and utilizes themcpcCLI tool for dynamic schema fetching. - [EXTERNAL_DOWNLOADS]: The documentation recommends the installation of the
@apify/mcpcpackage from the npm registry, which is the official tool for interacting with the Apify platform. - [DATA_EXFILTRATION]: The skill performs network operations targeting
api.apify.comandmcp.apify.com. These requests are necessary to trigger scraper runs and retrieve results, and they are restricted to the official vendor domains. - [PROMPT_INJECTION]: Potential surface for indirect prompt injection via scraped data.
- Ingestion points: Scraped content from third-party websites (e.g., social media posts, reviews) is retrieved via
api.apify.comand processed inreference/scripts/run_actor.js. - Boundary markers: No specific delimiters or instructions are used to separate untrusted scraped data from agent instructions in the output.
- Capability inventory: The skill can write to the local file system (
writeFileSync) and execute shell commands. - Sanitization: Content is truncated for display but not explicitly sanitized to remove potential instruction sequences embedded in scraped text.
Audit Metadata