The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The scripts/check-updates.py utility fetches metadata from the official npm registry at registry.npmjs.org to compare the tracked version with the latest available release. This is a standard documentation maintenance task targeting a whitelisted domain.
[COMMAND_EXECUTION]: The SKILL.md and references/00-overview.md files provide standard installation and startup commands for the n8n platform, including npx n8n, npm install n8n -g, and docker run. These are intended for user-guided setup of the official software.
[INDIRECT_PROMPT_INJECTION]: The skill documents an automation platform that is designed to ingest untrusted data from external sources.
Ingestion points: External data enters the context through references/02-triggers-and-webhooks.md (Webhook Triggers) and references/06-http-request-and-apis.md (HTTP Request responses).
Boundary markers: The skill documentation mentions the use of System Messages in references/09-ai-agents-and-tools.md to guide agent behavior when processing input.
Capability inventory: The platform supports arbitrary code execution via the Code node (references/05-code-node.md), network requests (references/06-http-request-and-apis.md), and file system operations.
Sanitization: As documented in references/07-credentials-and-security.md, the platform provides built-in SSRF (Server-Side Request Forgery) protection, credential encryption at rest, and domain restrictions to mitigate the risks of processing untrusted data.

n8n