The Agent Skills Directory

[COMMAND_EXECUTION]: The skill is designed to execute shell commands for environment management (startup/shutdown) and running test suites. It follows instructions found in internal project files like ac-spec.md and lessons.md to determine the specific commands to run.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface for indirect prompt injection.
Ingestion points: It reads untrusted data from eo-doc/changes/<change-id>/change.md, Git commit diffs, and ../eo-shared/lessons.md to define its testing strategy.
Boundary markers: The instructions do not specify any delimiters or warnings to ignore embedded instructions within these data sources.
Capability inventory: The skill has the capability to execute arbitrary test commands and write files to the local file system.
Sanitization: There is no mention of sanitization or validation of the content extracted from these files before it is used to influence agent behavior or command generation.
[DYNAMIC_EXECUTION]: The skill dynamically generates and saves new test files to the repository and subsequently executes them. While this is part of its core functionality for covering testing gaps, the execution of generated code based on external inputs (ACs and code diffs) presents an inherent risk.

eo-test