The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches a deployment configuration (docker-compose.yml) from the official ReportPortal GitHub repository. ReportPortal is a well-known automated testing platform, and the reference is documented neutrally.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it processes untrusted test result data that could be influenced by external contributors or malicious code under test.
Ingestion points: Test execution data including JUnit XML, coverage JSON, and timing JSON are parsed by the push-test-metrics.ts script and Allure report generators.
Boundary markers: The skill does not implement delimiters or specific warnings to ignore instructions embedded within the ingested test data.
Capability inventory: The skill includes network communication with InfluxDB and ReportPortal, as well as shell command execution through npx and docker compose.
Sanitization: There is no evidence of sanitization or schema validation for the contents of the result files before processing.
[COMMAND_EXECUTION]: The instructions include various commands for report generation (allure generate), serving data (allure serve), and managing container environments (docker compose up). These are standard for the skill's intended purpose.

qa-dashboard