The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content from external test files during the promotion and merging workflow, which could harbor malicious instructions.
Ingestion points: Test specification files located in e2e/tests/handover/ and e2e/tests/regression/ are read and parsed to identify test.describe blocks.
Boundary markers: There are no instructions to the agent to use delimiters or to ignore embedded natural language instructions within the code or comments of the files being processed.
Capability inventory: The agent is instructed to perform file system operations (moving/renaming files), file content editing (merging code blocks), file deletion (rm), and shell command execution (npx playwright test).
Sanitization: The workflow lacks explicit sanitization or validation of the file content before the agent interprets or modifies it.
[COMMAND_EXECUTION]: The skill directs the agent to execute shell commands using inputs derived from the filesystem.
Evidence: The command npx playwright test --project="chromium:regression" {feature}.spec.ts relies on the {feature} variable. If the file name or ticket prefix is maliciously crafted, it could potentially lead to command injection if the agent does not properly escape the arguments.

promote-handover-test