The Agent Skills Directory

[DATA_EXFILTRATION]: The skill accesses the APIFY_TOKEN stored in a local .env file and transmits it to the official Apify API (api.apify.com) and MCP server (mcp.apify.com) for service authentication. This is standard operational behavior for interacting with the Apify platform.
[COMMAND_EXECUTION]: The skill requires the use of the mcpc CLI tool and a provided Node.js script to orchestrate data collection and processing tasks. These operations are executed within the agent's environment as instructed by the workflow.
[PROMPT_INJECTION]: The skill is designed to ingest and interpret content from external platforms like Instagram, Facebook, and TikTok, which represents an indirect prompt injection surface.
Ingestion points: Untrusted content is fetched from social media platforms via various Apify Actors as described in SKILL.md and processed by the run_actor.js script.
Boundary markers: The instructions do not define explicit delimiters or instructions to the agent to disregard potential commands embedded within the retrieved social media data.
Capability inventory: The skill has the capability to write files to the local system and execute CLI commands based on the data retrieved.
Sanitization: The provided Node.js script performs basic string truncation and CSV escaping for data formatting, but it does not include measures to sanitize the content for potential prompt injection attacks.

apify-content-analytics