The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the notion-cli command-line utility for managing Notion pages, databases, and blocks. This behavior is expected and aligned with the skill's stated purpose.- [CREDENTIALS_UNSAFE]: The skill documentation correctly identifies the need for a NOTION_TOKEN or the use of notion-cli auth. It does not include or encourage the use of hardcoded secrets, following security best practices for credential management.- [PROMPT_INJECTION]: The skill interacts with untrusted data from Notion workspaces, which presents a surface for indirect prompt injection.
Ingestion points: Data is retrieved using notion-cli page get, notion-cli page search, notion-cli db query, and notion-cli block list as described in SKILL.md.
Boundary markers: No specific boundary markers or 'ignore' instructions are provided to the agent to distinguish between system instructions and retrieved Notion content.
Capability inventory: The skill allows for data modification via create, update, append, and delete commands.
Sanitization: No explicit sanitization or validation of retrieved content is implemented within the skill instructions.

c-notion