The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's instructions and execution logic involve running shell commands to set up the environment and execute the script. Specifically, it uses npm install and npx -y tsx to manage dependencies and run the TypeScript source code.
[EXTERNAL_DOWNLOADS]: The npm install command downloads third-party packages from the public npm registry. While the provided package.json only lists tsx, the installation process inherently fetches external code.
[DATA_EXPOSURE]: The script scripts/main.ts accesses a sensitive credential file located at ~/.config/notion/api_key. This access is intended for its primary purpose of authenticating with the Notion API.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted local Markdown files which serves as an attack surface for indirect prompt injection.
Ingestion points: The scripts/main.ts script reads markdown files from paths provided as command-line arguments.
Boundary markers: None identified; the agent is not instructed to use delimiters or ignore instructions within the files.
Capability inventory: The skill can perform network operations via the Notion API and execute shell commands through the setup process.
Sanitization: There is no evidence of sanitization or filtering of the markdown content before it is processed or sent to the API.

notion-md-upload-skill