docs(skills): exclude caveman terse style from commit and MR/PR artifacts

[paolomainardi]

🤖 This was written by an AI agent on behalf of @paolomainardi.

Summary

When a session runs with an active terse output style (the caveman plugin's CAVEMAN MODE ACTIVE reminder), that compression can leak into durable, human-facing artifacts: commit messages, MR/PR titles and descriptions, comments, and reviews. This change makes the carve-out explicit and consistent across our artifact-producing skills.

• gh and glab: merge the two separate "write in plain prose" and caveman-override paragraphs into a single section per skill, and state that an active terse style does not apply to titles, descriptions, comments, notes, or reviews.
• sf-commit-convention: add a "Write in plain prose, not caveman" section covering commit subject and body and MR/PR titles. This is the skill loaded before every commit, so it is the central guard.
• CHANGELOG.md: one consolidated entry under [Unreleased].

Rationale

The upstream caveman skill only guarantees a narrow carve-out in its Boundaries section (Code/commits/PRs: write normal.). It does not cover MR/PR descriptions, titles, issue descriptions, comments, or any GitLab vocabulary. That gap is acknowledged upstream in JuliusBrussee/caveman#484 and PR JuliusBrussee/caveman#483, both still open and unmerged. There is no config knob to extend the carve-out, so the established pattern is per-skill boundary statements (see the upstream caveman-commit and caveman-review skills). This change follows that pattern.

Scope

Four files only. A repo-wide markdown formatter run touched many unrelated files in the working tree; those were intentionally left unstaged and are not part of this PR.

Closes #91

[sparkfabrik-ai-bot]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

⏱️ Estimated effort to review: 1 🔵⚪⚪⚪⚪

🧪 No relevant tests

🔒 No security concerns identified

⚡ Recommended focus areas for review Dense Paragraph The new single-paragraph block in the gh skill is very long and dense, merging what were previously two separate, clearly delineated paragraphs. This may reduce readability for the AI consuming the skill, potentially making the instruction less effective than the original two-paragraph format used in the glab skill. **Write in plain, professional prose.** Issue and pull request titles and descriptions, comments and reviews, and commit messages must always be written in normal, complete, well-structured English: full sentences, articles, and proper markdown. They are durable, outward-facing documents that other people read and that GitHub renders in its UI; compression hurts readability and looks unprofessional. This rule **overrides any active terse output style for the duration of writing these artifacts.** A session-level style may inject a per-turn reminder such as `CAVEMAN MODE ACTIVE` (drop articles, fragments OK, short synonyms); that reminder applies to your conversational replies, not to the artifacts. When you compose an issue/PR title or body, a comment, a review, or a commit message, write full prose regardless of the active style, then resume the terse conversational style for your surrounding chat replies. Do not run any command to toggle the style off; just write the artifact in plain prose regardless of what reminder is present.

[sparkfabrik-ai-bot]

PR Code Suggestions ✨

No code suggestions found for the PR.