The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill defines a surface where the agent ingests external instructions and runbooks from the repository. This is an attack surface where malicious content within those files could attempt to override the agent's verification logic.
Ingestion points: The skill explicitly instructs the agent to read repository instructions and the task's QA Runbook (SKILL.md).
Boundary markers: No specific boundary markers or warnings to disregard instructions within the external data are present.
Capability inventory: The skill possesses the ability to execute shell commands for testing, including php artisan and npm scripts (SKILL.md).
Sanitization: There is no evidence of sanitization or validation logic for the ingested documentation content.
[COMMAND_EXECUTION]: The skill provides the agent with specific shell commands to execute testing suites, which is appropriate for its stated purpose as a verification tool.
Evidence: Commands mentioned include php artisan test:isolated and npm run e2e:myhub:live, as well as Cypress-based tests (SKILL.md).

sociuu-verify