The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes a local Python helper script (scripts/query-tasks.py) to interface with the Notion REST API for task synchronization. This execution occurs when the user triggers the refresh or run commands.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It retrieves task titles, descriptions, and discussion logs from a Notion database and includes them in the prompts ("kickoff briefs") provided to agents.
Ingestion points: Data from the Notion Tasks database is ingested via scripts/query-tasks.py and the start command.
Boundary markers: The skill does not implement boundary markers or instructions for the agent to disregard potential injection content within the Notion-sourced data.
Capability inventory: The agent has access to sensitive tools including Bash shell execution, python3, npx, and direct Notion workspace manipulation tools.
Sanitization: No sanitization or validation is applied to the free-text content retrieved from Notion before it is used to build prompts.
[EXTERNAL_DOWNLOADS]: The skill instructs the installation of the official Notion MCP server (@notionhq/notion-mcp-server) via npx. As this targets a well-known service from a recognized organization, this behavior is documented neutrally as intended functionality.

agency-os