The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of external CLI tools from public registries: notion-cli (NPM) and notion-cli-py (PyPI).
[COMMAND_EXECUTION]: All Notion operations are performed by executing local CLI commands (e.g., notion-cli page get, notion-cli db query).
[CREDENTIALS_UNSAFE]: The documentation follows best practices for secret management by instructing users to store the NOTION_API_KEY in environment variables rather than hardcoding it.
[DATA_EXPOSURE]: The skill is designed to read and write content within a user's Notion workspace, which is the intended primary purpose of the integration.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests untrusted content from Notion pages and databases which could contain malicious instructions.
Ingestion points: Content is retrieved from Notion pages via notion-cli page get (SKILL.md).
Boundary markers: None identified in the provided instructions to isolate retrieved content from the agent's control flow.
Capability inventory: The skill can execute shell commands via notion-cli and write to the filesystem or network depending on the CLI's implementation (SKILL.md).
Sanitization: No explicit sanitization or validation of the content retrieved from Notion is described.

notion-skill