The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted source code (Cypress and Selenium tests) and converts it into executable Playwright tests. An attacker could craft a source test file containing malicious payloads designed to be preserved or activated during conversion, leading to code execution during the verification step.
Ingestion points: Reads files from cypress/ directory and Selenium test files (.py) as described in Step 1 of SKILL.md.
Boundary markers: None. The instructions do not specify any delimiters or warnings to ignore instructions embedded within the source code.
Capability inventory: The skill possesses the ability to write files to the local system and execute commands via the terminal (npx playwright test).
Sanitization: The skill lacks explicit sanitization or validation logic to ensure that converted code does not contain malicious shell escapes or unauthorized JS commands.
[COMMAND_EXECUTION]: The skill is designed to execute shell commands to verify the success of the migration.
Evidence: Step 7 in SKILL.md instructs the agent to run npx playwright test <converted-file> --reporter=list on the newly generated code.

migrate