The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes concept documentation pages to extract code examples for test generation. This ingestion of untrusted external content represents a vulnerability surface where malicious instructions embedded in documentation could potentially influence agent behavior.
Ingestion points: Documentation content scanned in Phase 1 (Code Example Extraction).
Boundary markers: None specified in the instructions to separate documentation content from agent instructions.
Capability inventory: File writing, directory creation, and shell command execution (npm test).
Sanitization: No explicit sanitization or validation of the extracted code examples is described.
[DYNAMIC_EXECUTION]: Pattern 16 (Strict Mode Behavior) includes a template using the eval() function to test undeclared variable errors. While used in a restricted, educational context within the test code, eval() is a dynamic execution sink.
[COMMAND_EXECUTION]: The skill provides instructions for the agent to execute shell commands via npm test and its variants to verify the generated tests, which is consistent with the skill's primary purpose.

test-writer