apify-ecommerce
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute a local Node.js script (
run_actor.js) which uses standard Node.js built-ins to interact with the Apify API. The script accepts JSON input and command-line arguments that dictate how the data is processed and stored. - [DATA_EXFİLTRATION]: The skill uses a sensitive API token (
APİFY_TOKEN) from the local filesystem (~/.claude/.env). This token is sent toapi.apify.com, which is a well-known and legitimate service for web scraping, but users should be aware of the data flow to this external provider. - [INDIRECT_PROMPT_INJECTION]: The skill fetches content from external, untrusted sources such as product descriptions and customer reviews. This data is then processed and summarized by the agent, creating a surface for potential indirect prompt injection.
- Ingestion points: External data from e-commerce platforms is ingested via the Apify API and processed in
run_actor.jsbefore being presented to the agent. - Boundary markers: The instructions do not define specific delimiters or "ignore" instructions for the agent when analyzing the scraped data.
- Capability inventory: The agent has capabilities for file system access, network requests, and command execution.
- Sanitization: While the script implements basic protections against CSV formula injection, it does not sanitize content for prompt injection patterns.
Audit Metadata