The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and analyze user-controlled source code and test files to perform its primary tasks. This creates a surface for indirect prompt injection if the analyzed files contain malicious instructions.
Ingestion points: Files are read in the initialization steps of the reviewing, writing, and debugging workflows (e.g., workflows/reviewing/steps/step-01-init.md).
Boundary markers: There are no instructions for the agent to use delimiters or specific ignore-rules when processing file content.
Capability inventory: The agent can execute shell commands (npm test), read files, and write files to the local system.
Sanitization: No sanitization is performed on the ingested code before processing.
[COMMAND_EXECUTION]: The skill defines workflows that use local shell commands to execute test suites and manage test output.
Evidence: SKILL.md and the workflows/ directory contain commands such as npm test, tail, grep, and rm for running tests and handling log files in /tmp/.
[EXTERNAL_DOWNLOADS]: The skill references and suggests installing several standard development and testing packages from the NPM registry.
Evidence: Documentation in the references/ directory mentions packages like mongodb-memory-server, pg-mem, ioredis, and various Jest-related plugins and reporters.

typescript-unit-testing