The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data to guide its code generation and testing actions.
Ingestion points: The skill accepts bug reports, code diffs, and API change descriptions as inputs (SKILL.md).
Boundary markers: No specific delimiters or safety instructions are defined to separate untrusted input data from the agent's core instructions.
Capability inventory: The skill has the capability to write new test files and execute shell commands via bun test (SKILL.md).
Sanitization: The instructions do not specify any validation or sanitization of the input data before it is interpolated into the testing workflow.
[COMMAND_EXECUTION]: The skill executes local shell commands as part of its testing and verification workflow.
Evidence: The skill instructs the agent to run bun test, bun run typecheck, and other project-specific test commands (SKILL.md).
[DYNAMIC_EXECUTION]: The skill generates and then executes test scripts based on its analysis of the code and bug reports.
Evidence: The workflow involves creating or modifying test files and fixtures which are then executed by the bun test runner (SKILL.md).

ai-regression-testing