The Agent Skills Directory

[DATA_EXPOSURE]: The skill implements proactive data protection by providing logic to scrub sensitive information, such as emails, credit card numbers, and social security numbers, from Sentry events using the beforeSend hook.
[INDIRECT_PROMPT_INJECTION]: The skill audits data retrieved from Sentry APIs to ensure compliance with organizational standards, representing a managed data ingestion surface.
Ingestion points: getProjects() and getProjectAlerts() functions in references/audit-and-compliance.md.
Boundary markers: Not explicitly defined in the provided audit logic.
Capability inventory: Uses Read, Write, Edit, and Grep tools for configuration management.
Sanitization: Includes regex-based validation (validateProjectName) to enforce naming standards and filter non-compliant input.
[SAFE]: All network operations are directed at the official Sentry API (sentry.io), and sensitive credentials like SENTRY_TOKEN are handled as environment variable placeholders.

sentry-policy-guardrails