Install

The Everr CLI helps AI agents and humans move quickly from CI/CD signal to action.

It is designed so teams can inspect failures, logs, and branch status from the current repository context without switching tools.

Install

Install the latest Everr CLI build:

curl -fsSL https://everr.dev/install.sh | sh

The installer places everr in ~/.local/bin and starts setup when an interactive terminal is available.

Commands

• everr status
• everr grep --job-name <job> --step-number <n> --pattern <text>
• everr install
• everr slowest-tests (repo-wide for non-suite tests by default; use --branch <name> to scope it)
• everr slowest-jobs (repo-wide by default; use --branch <name> to scope it)
• everr wait-pipeline (waits until local HEAD commit appears in runs; use --commit <sha> to target a specific commit; exits non-zero if any completed run finishes with failure)
• everr runs list
• everr runs show --trace-id <trace_id>
• everr runs logs --trace-id <trace_id> --job-name <job> --step-number <n>

everr status returns a failure-first JSON payload. When status.failures[i].logsArgs is present you can call everr runs logs directly; otherwise use everr runs show to discover the failing step.

everr grep searches failing step logs for a literal, case-insensitive pattern. Use --job-name together with --step-number to target one specific failing step. By default it searches the current repository over the last 7 days and excludes the currently checked out branch so you can quickly answer whether the same failure happened elsewhere.

Collection-style commands accept --limit <n> and --offset <n> for pagination. everr runs list also continues to support --page <n>.