The Agent Skills Directory

[DATA_EXFILTRATION]: The skill reads an authentication token from the local file ~/.claude/.env and transmits it to api.apify.com. This is consistent with the skill's purpose to authenticate requests to the Apify API.
[CREDENTIALS_UNSAFE]: The run_actor.js script transmits the authentication token as a query parameter in the API request URLs. This practice is less secure than using Authorization headers, as query strings may be captured in web server logs or command history.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it retrieves and processes untrusted data from third-party e-commerce marketplaces.
Ingestion points: Data such as product descriptions and reviews are fetched from external marketplaces via the Apify API in run_actor.js.
Boundary markers: The skill does not employ delimiters or specific instructions to isolate external scraped content from the agent's logic.
Capability inventory: The skill can write files to the local system (writeFileSync) and make network requests (fetch).
Sanitization: No validation or sanitization is performed on the scraped data before it is returned to the agent's context.

apify-ecommerce