From 97f727f468927c79a1f9a4b6fd0c6359a1778bf2 Mon Sep 17 00:00:00 2001 From: Alan Silva Date: Fri, 13 Nov 2026 12:45:23 +0000 Subject: [PATCH] Skill: always wait on a started run with pipelines wait Agents started runs and moved on, then missed failures. The skill now says to follow every run with a background pipelines wait, read the result when it ends, and act on it instead of polling or promising a report. --- internal/author/skill/SKILL.md | 10 +++++++++- 1 file changed, 9 insertions(+), 1 deletion(-) diff --git a/internal/author/skill/SKILL.md b/internal/author/skill/SKILL.md index 760731b8..d8871dc0 100644 --- a/internal/author/skill/SKILL.md +++ b/internal/author/skill/SKILL.md @@ -13,7 +13,15 @@ task with the least ceremony, and keep work small enough to verify. `pipelines list` shows the registered workflows; `pipelines runs` shows history. Start a run with `pipelines run ""`, and inspect it with -`pipelines inspect `. `pipelines wait ` blocks until it finishes. +`pipelines inspect `. + +After starting a run, always follow it with `pipelines wait ` until it +finishes; never start a run and move on without waiting. Run the wait as a +background command (one per run, or one loop over several ids) so the session is +notified the moment the run ends, then read its result with +`pipelines inspect --json` and act on it: land, retry, or fix and report. +Do not poll with repeated `pipelines runs` calls, and do not tell the user a run +is done, or will be reported, without a wait covering it. A failed run keeps its worktree and commits. Read `pipelines inspect --json` before deciding: resume the failed stage (`pipelines retry `), fix directly,