The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses pnpm to start and stop a disposable site (pnpm ux:site:start and pnpm ux:site:stop). These are local project-specific commands used to manage the lifecycle of a test environment.
[INDIRECT_PROMPT_INJECTION]: The coordinator agent processes untrusted data from a separate tester agent, creating a vulnerability surface where instructions embedded in the test logs could influence the coordinator's behavior.
Ingestion points: The skill reads action logs, screenshots, and reports produced by a separate tester agent, as well as journey definitions from acceptance/journeys/.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded commands within the tester's report or journey files.
Capability inventory: The skill can execute shell commands via pnpm and perform direct database inspections.
Sanitization: There is no mention of sanitizing or validating the contents of the tester's logs or action history before processing.

ux-acceptance-coordinator