diff --git a/.claude/skills/pr/SKILL.md b/.claude/skills/pr/SKILL.md index a6e1c0888bf..ce31bfeba0a 100644 --- a/.claude/skills/pr/SKILL.md +++ b/.claude/skills/pr/SKILL.md @@ -17,14 +17,23 @@ the PR body.** where `` comes from `gh api user --jq .login`. 2. Gather context: `git status`, `git log --oneline main..HEAD`, `git diff main...HEAD --stat`, check for remote tracking branch. -3. Sanity-build the targets we care about. Fix any build errors before opening - the PR: - - `make px4_fmu-v6x` — hardware target - - `make px4_sitl` — simulation +3. Sanity-build **one** target the change can actually affect — a board for + firmware changes, `px4_sitl` for POSIX-only or simulation changes. Skip the + build entirely when the diff cannot reach any target (submodule pointer + bumps, docs, ROMFS). Fix any build errors before opening the PR. 4. PR **title:** `type(scope): description` — under 72 chars, covers the overall change across all commits. This becomes the squash-merge commit message. -5. PR **body:** concise and terse — do not restate what the diff already shows (no file-changed lists, no code snippets that reproduce the diff). Use exactly three sections, in order: `## Summary`, `## Problem`, `## Solution`. If the PR closes a GitHub issue, the first line of `## Summary` must be `fixes #`, then a blank line, then the summary text. No `## Test plan` section, no boilerplate, no Claude attribution. Use markdown (links, code blocks, lists) only when warranted. Never state testing that did not happen: ask the user what they actually ran beyond the builds in step 3, report exactly that, and say so plainly when something is untested. +5. PR **body:** as short as it can be while still landing the point — a + reviewer should take it in at a glance, and a long description is one + nobody reads. Exactly three sections, in order: `## Summary`, `## Problem`, + `## Solution`, a sentence or two each. Do not restate the diff (no + file-changed lists, no code snippets), do not mention CI, and do not repeat + what the title already says. If the PR closes an issue, the first line of + `## Summary` is `fixes #`, then a blank line, then the summary. No + `## Test plan` section, no boilerplate, no Claude attribution. Never state + testing that did not happen: ask the user what they actually ran, report + exactly that, and say plainly when something is untested. 6. Push with `-u` if needed, then `gh pr create`. Default base is `main` unless user says otherwise.