The Agent Skills Directory

[PROMPT_INJECTION]: The skill defines patterns for fetching and processing data from external databases, which creates an indirect prompt injection surface when combined with technical capabilities.
Ingestion points: Processes organization-specific data fetched via database queries as seen in the dailyDigest example in SKILL.md.
Boundary markers: Does not include explicit delimiters or instructions for the agent to treat external data as untrusted or to ignore embedded commands.
Capability inventory: Utilizes Bash, Read, and Grep for managing backend configurations and system files.
Sanitization: No specific data sanitization or validation steps are prescribed for the agent to perform on data retrieved from the database before it is included in the execution context.
[EXTERNAL_DOWNLOADS]: The skill references documentation and implementation guides from well-known services and organizations, including Vercel, Inngest, and OpenAI. These references are used for architectural grounding and are sourced from official domains.
[COMMAND_EXECUTION]: The skill involves the configuration and management of deployment-related files (such as vercel.json) and backend script handlers, requiring standard command-line tools for execution.

cron-scheduling