The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to interact with external API endpoints to perform testing. This creates a surface where the agent could ingest data containing malicious instructions from an attacker-controlled API.
Ingestion points: API response data retrieved through fetch calls as described in the testing examples within SKILL.md.
Boundary markers: No explicit delimiters are specified in the instructions to prevent the agent from interpreting API content as instructions.
Capability inventory: The agent has capabilities for network operations and shell command execution via testing frameworks like Playwright.
Sanitization: The instructions focus on testing the target API's sanitization but do not explicitly define sanitization for the agent's own processing of the results.
[COMMAND_EXECUTION]: The workflow requires the use of command-line tools and automation frameworks (e.g., Playwright, k6) which involve executing scripts and commands in the local environment. This is consistent with the skill's primary purpose.
[EXTERNAL_DOWNLOADS]: The skill mentions the use of standard, well-known industry packages and tools such as @playwright/test and k6 for automated testing and performance monitoring.

testing-api-tester