skills/buildkite/skills/buildkite-cli/Gen Agent Trust Hub

buildkite-cli

Pass

Audited by Gen Agent Trust Hub on Apr 30, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: Retrieves the Buildkite CLI (bk) and associated installation scripts from official Buildkite GitHub repositories and Homebrew taps, which are verified vendor sources.
  • [COMMAND_EXECUTION]: Utilizes the bk command-line utility to perform operations such as triggering builds, viewing logs, and managing pipelines. The skill includes specific warnings against passing sensitive values as literal strings to avoid exposure in shell history. It also provides standard instructions for generating shell completions, which may involve writing to system paths like /etc/bash_completion.d/.
  • [PROMPT_INJECTION]: Identifies an indirect prompt injection surface where the agent processes external data such as build logs (bk job log) or artifacts (bk artifact download). While the agent possesses capabilities like command execution and secret management, there is no evidence of boundary bypasses or malicious instructions intended to exploit these surfaces.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 30, 2026, 01:52 AM
Security Audit — agent-trust-hub — buildkite-cli