The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from Instagram profiles, hashtags, and locations, which could contain malicious instructions designed to influence the agent's behavior.
Ingestion points: Instagram data retrieved via the startUrls input parameter (SKILL.md).
Boundary markers: The instructions lack specific delimiters or instructions for the agent to ignore embedded commands within the scraped content.
Capability inventory: The skill environment supports shell command execution (node) and network requests (curl) to the Apify API (SKILL.md).
Sanitization: No sanitization or validation logic is defined for the external data before it is integrated into the agent's context.
[DYNAMIC_EXECUTION]: The customMapFunction parameter allows the user or agent to pass a JavaScript string to the Apify actor. Although this code executes on the remote Apify platform rather than the local agent environment, it creates a vector for remote code execution if the input is improperly handled.
[COMMAND_EXECUTION]: The documentation provides explicit instructions for the agent to execute shell commands using node to run local scripts and curl to perform HTTP POST requests to the Apify REST API.
[EXTERNAL_DOWNLOADS]: The skill references a script named scripts/run_actor.js as the recommended way to execute the actor. This script is not included in the skill's file list, making it an unverifiable external dependency.

finding-restaurant-brands-on-instagram