The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from GitHub pull requests and issues, which could contain malicious content designed to influence the agent's summarization or reporting logic.
Ingestion points: Data is read via gh pr list, gh issue list, and local files such as output/articles/token-report-*.md.
Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted external content from the skill's own reporting templates.
Capability inventory: The skill has the ability to write to the public docs/status.md file, send system notifications, and make external network requests via the Resend API.
Sanitization: The skill extracts specific fields using regex but does not perform explicit sanitization of the narrative text content from PRs or issues.
[EXTERNAL_DOWNLOADS]: The skill communicates with the Resend API to send email briefings.
Source/Destination: https://api.resend.com/emails.
Mechanism: Utilizes the platform's secure curl implementation to handle the RESEND_API_KEY for communication with the Resend service.
Note: This activity targets a well-known service and is restricted to the skill's documented primary function.

heartbeat