The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external data from build logs and test summaries which may contain untrusted content.
Ingestion points: Build logs retrieved via bk job log, structured event data from bk preflight --json, and test engine failure reports.
Boundary markers: The instructions do not define specific delimiters or instructions to ignore embedded content within logs.
Capability inventory: The skill utilizes shell command execution through bk, git, and jq tools.
Sanitization: No explicit sanitization or filtering is prescribed for the content of build logs or job outputs before the agent interprets them.
[COMMAND_EXECUTION]: The skill's primary functionality is implemented through the execution of shell commands.
Evidence: The skill provides numerous examples for using bk preflight, bk build view, bk job list, and bk job log to interact with CI infrastructure.

buildkite-preflight