The Agent Skills Directory

[PROMPT_INJECTION]: The skill processes untrusted data from external Apify Actors which introduces a risk of indirect prompt injection.
Ingestion points: Data is fetched from the Apify API (api.apify.com) as seen in reference/scripts/run_actor.js.
Boundary markers: The instructions in SKILL.md do not specify the use of delimiters to isolate the scraped data from system instructions.
Capability inventory: The agent can execute shell commands and write files, providing a potential impact path for injected content.
Sanitization: The run_actor.js script lacks sanitization logic to strip instructions from the actor output.
Remediation: Use clear delimiters and instructions to ignore embedded commands in external content.
[COMMAND_EXECUTION]: The skill uses the mcpc CLI and a local script for scraping functions.
Details: These actions are required for the skill's primary purpose and are executed within a standard development environment.
[EXTERNAL_DOWNLOADS]: The skill requires downloading the @apify/mcpc tool from NPM.
Details: This is a standard dependency for Apify services and is obtained from a well-known registry.

apify-lead-generation