The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill instructs the agent to download and execute scripts directly into the shell using curl -fsSL https://ntn.dev | bash. This pattern allows a remote server to execute arbitrary commands with the user's privileges, and the content of the script can be changed by the remote host at any time without notice.
[CREDENTIALS_UNSAFE]: The skill requires the use of sensitive Notion API integration tokens (NOTION_API_KEY, NOTION_API_TOKEN) and provides instructions to store them in local files such as .env and ~/.config/notion/auth.json. The agent is also instructed to bypass standard login procedures and manage these secrets directly, increasing the risk of exposure.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by processing untrusted data from external sources.
Ingestion points: The skill reads page content and database queries from the Notion API (/v1/pages/{id}/markdown, /v1/data_sources/{id}/query).
Boundary markers: None. There are no instructions to the agent to ignore or delimit instructions found within the Notion data.
Capability inventory: The skill has the ability to execute shell commands (via ntn CLI and curl), write files (via the export script logic), and make arbitrary network requests.
Sanitization: No sanitization or validation of the ingested Notion content is performed before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill performs several external downloads and installations from non-whitelisted sources:
Downloads and executes an installation script from https://ntn.dev.
Installs the ntn package globally via npm install --global ntn from the public NPM registry.
Fetches data and markdown content from api.notion.com.

notion