The Agent Skills Directory

[SAFE]: The skill is a legitimate developer toolset for the Apify platform. It follows industry security standards and provides explicit mitigations for platform-specific risks.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the official apify-cli via standard package managers (npm, homebrew). It also manages project-level dependencies via package-lock.json and requirements.txt, adhering to secure supply-chain practices.
[COMMAND_EXECUTION]: Instructions involve using local CLI tools for actor development, project scaffolding (apify create), local testing (apify run), and cloud deployment (apify push). All commands are standard developer workflows for the Apify ecosystem.
[INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface where Actors process untrusted web content. It implements a mitigation strategy by providing mandatory security instructions:
Ingestion points: External websites scraped by Actors (as described in SKILL.md).
Boundary markers: The skill provides a dedicated 'Security' section with explicit rules for data handling.
Capability inventory: Actors have capabilities for network scraping and data persistence; the developer tool allows local code execution via apify run.
Sanitization: The instructions strictly mandate sanitizing all crawled data and prohibit passing raw content to shell commands or eval() functions.
[CREDENTIALS_SAFE]: The skill provides clear instructions for handling the APIFY_TOKEN, specifically warning against exposing secrets in process listings or shell history.

apify-actor-development