The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands to interact with the Apify platform, specifically calling the mcpc CLI to fetch Actor schemas and running a Node.js script to perform data scraping.
[EXTERNAL_DOWNLOADS]: Requires the user to install @apify/mcpc via npm, which is the official Model Context Protocol client for Apify, a well-known technology provider.
[DATA_EXFILTRATION]: The skill accesses a sensitive APIFY_TOKEN stored in a local .env file. This token is used to authenticate POST and GET requests to api.apify.com for running Actors and retrieving results, which is consistent with the skill's documented functionality.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it ingests untrusted data from various external websites (Instagram, TikTok, Google Maps) and presents it to the agent.
Ingestion points: Scraped data is downloaded from Apify datasets via api.apify.com in run_actor.js (functions downloadResults and displayQuickAnswer).
Boundary markers: No explicit delimiters or instructions are used to separate scraped content from agent instructions.
Capability inventory: The skill can perform network requests (fetch in run_actor.js) and write results to the local filesystem (writeFileSync).
Sanitization: The script performs basic CSV escaping and truncates long strings for display, but does not sanitize content for potentially malicious instructions.

apify-lead-generation