The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a local binary named asana to execute commands such as status, list-tools, and call-tool for interacting with the Asana platform.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data including tasks, comments, and project details from Asana, which represents a potential surface for indirect prompt injection.
Ingestion points: Data is ingested through the asana call-tool command which fetches content from Asana's API (manifest.yaml, SKILL.md).
Boundary markers: The instructions do not specify explicit boundary markers or delimiters for the ingested content.
Capability inventory: The skill has the capability to write, update, and delete tasks, projects, and comments based on user-approved scopes (manifest.yaml).
Sanitization: The skill relies on the structured tool-calling interface of the asana CLI and the Model Context Protocol (MCP) server for data handling.
[EXTERNAL_DOWNLOADS]: The skill references and interacts with official Asana domains (mcp.asana.com and app.asana.com) for OAuth 2.0 authentication flows and API communication (connector_auth_config.yaml).

asana