The Agent Skills Directory

Indirect Prompt Injection Surface: The skill processes information from external sources such as issue descriptions and system logs to guide the debugging process.
Ingestion points: Data is gathered from issue reports and automated run logs via GitHub CLI commands in SKILL.md.
Capability inventory: The agent has capabilities to execute build commands, run tests, and perform version control operations (commit, push, PR creation).
Boundary markers: The instructions implement mandatory checkpoints and recommend delegating log analysis to specialized sub-agents, which serves to isolate and validate external input before it influences core logic.
Sanitization: The workflow relies on the agent's interpretation of logs and issues rather than direct execution of their contents.
Environment State Management: The utility scripts perform cleanup operations to ensure test reliability and isolation.
Evidence: The scripts run-test-repeatedly.sh and run-test-repeatedly.ps1 terminate specific running processes (such as test runners and service hosts) and clean up containerized resources using Docker commands.
Context: These actions are necessary for resetting the environment between test iterations, especially in complex distributed environments. The scripts include safety prompts for local users before performing impactful cleanup operations.

fix-flaky-test