The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from multiple external sources, which could contain malicious instructions intended to influence the agent's behavior.
Ingestion points: The workflow gathers data from user reports, support messages, app-store reviews, Sentry issues, and PR review comments (Phase 0).
Boundary markers: The skill instructs the agent to 'Normalize each into a common shape', providing a structural boundary for the raw data.
Capability inventory: The agent has access to gh issue create and test-playwright to persist data and execute tests.
Sanitization: The normalization step and the requirement to write 'durable, reproducible tickets' act as manual and logical filters for external content.
[COMMAND_EXECUTION]: The skill utilizes command-line tools to manage external state and perform verification.
Evidence: Uses gh issue list and gh issue create for issue tracking, and test-playwright for live fix verification (Phase 1 and Phase 5).
Context: These tools are standard for the intended development environment and are used to fulfill the primary purpose of the skill.

workflow-feedback-to-closure