The Agent Skills Directory

[DATA_EXFILTRATION]: The skill retrieves private communications and task data from Google Workspace and Todoist, which it then transmits to external WhatsApp and Slack endpoints. Although this is the primary purpose of the skill, it involves the movement of sensitive information to third-party services.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text from email threads, meeting notes, and task descriptions, creating an attack surface for indirect prompt injection. Ingestion points: Reads external content via gog (Gmail), todoist-cli, and mcporter (Granola). Boundary markers: The instructions do not define any specific delimiters or 'ignore' instructions for the external content before it is processed. Capability inventory: The skill has access to shell execution (Bash), file writing (Write), and network operations (curl). Sanitization: No sanitization, filtering, or instruction-scrubbing logic is applied to the retrieved data before compilation.

executive-digest