The Agent Skills Directory

[SAFE]: The skill's primary function is performance analysis of CI/CD pipelines. It uses the GitHub CLI (gh) and jq for read-only metadata retrieval regarding job and step durations.
[COMMAND_EXECUTION]: The included bash script scripts/ci-timings.sh executes shell commands to process CI timing data. The script follows secure coding practices, such as using set -euo pipefail and properly quoting variables.
[INDIRECT_PROMPT_INJECTION]: The skill involves reading external data such as CI logs and GitHub API responses. This represents a potential indirect prompt injection surface; however, the skill's instructions focus on quantitative analysis and timestamp processing, which does not involve executing instructions found within that data.
[SAFE]: The skill explicitly advises the user to maintain security-critical 'guard' configurations, such as concurrency groups and change detection, which were originally implemented to prevent production incidents.

ci-speedup