wait-for

name	wait-for
description	Block execution until a condition is met. Use when waiting for pipelines, merge trains, deployments, file creation, or any event that can be checked with a shell command.

Wait For

Poll a command until it exits 0 (condition met), returns the fail code (failed early), or the timeout is reached. Set the bash tool's timeout ~30s higher than --timeout as a safety net.

./scripts/wait-for.sh "<command>" [--timeout N] [--interval N] [--fail-code N]

• The command is a single quoted string, evaluated by the shell (pipes/quotes work).
• timeout and intervals are number of seconds
• --fail-code N: exit code meaning "failed" — stop and return N (its stderr is surfaced). Use it for checks that can detect failure before the timeout.
• Exit codes:
  • 0 condition met
  • 1 timed out
  • 64 usage error
  • N the command returned your --fail-code (propagated as-is).

./scripts/wait-for.sh "test -f /tmp/done.flag" --timeout 120 --interval 5
./scripts/wait-for.sh "curl -sf https://example.com/health" --timeout 60 --interval 5

Wait for a pipeline

Compose wait-for.sh with pipeline-succeeded.sh (exits 0 succeeded, 1 running, 2 failed — failed jobs on stderr).

Run from the git repo (so glab finds the project) and use absolute paths (the scripts live in the skill folder). wait-for.sh passes its working directory through to the check, satisfying both:

SKILL="$HOME/.pi/agent/skills/wait-for/scripts"

# latest pipeline for the current branch
"$SKILL/wait-for.sh" "$SKILL/pipeline-succeeded.sh" \
  --timeout 600 --interval 30 --fail-code 2

# a specific pipeline (URL trailing number, or bare ID)
"$SKILL/wait-for.sh" "$SKILL/pipeline-succeeded.sh <url_or_id>" \
  --timeout 600 --interval 30 --fail-code 2

Result: exit 0 = succeeded, 2 = a required job failed (jobs listed on stderr), 1 = timed out.