The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides numerous shell command templates using curl and the apify CLI to diagnose network egress and interact with Apify Actors. These commands are intended for technical troubleshooting but rely on user-supplied variables like DOMAIN and HOST.- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to scrape and analyze data from external, untrusted websites. This presents a potential surface for indirect prompt injection, although the skill focuses on technical metadata (HTTP headers, status codes) rather than natural language interpretation. Ingestion points: External URLs scraped via Apify Actors. Boundary markers: None explicitly defined. Capability inventory: Access to shell commands and vendor Actor execution. Sanitization: Focuses on specific technical field extraction.- [DYNAMIC_EXECUTION]: The skill provides templates for preNavigationHooks and pageFunction inputs. These contain JavaScript code that is executed within the runtime environment of the Apify Actors to customize request behavior and data extraction.- [PRIVILEGE_ESCALATION]: The documentation notes that specific Actors require 'full account access' permissions to be approved in the Apify Console. This is a standard platform requirement for the described diagnostic functionality.- [SAFE]: All external services referenced (e.g., Cloudflare, Google DNS, ip-api.com) are well-known diagnostic tools or vendor-controlled resources (Apify platform).

apify-blocked-scrape-triage