What does this skill do, and when should you use it?
This skill structures raw output from changelog-scan into a polished release notes draft. It follows the exact structure and rules from Grok's draft-release-notes skill. The output is a RELEASE_NOTES_DRAFT.md file or a clear section for the loop to persist. It flags breaking changes and security issues at the top and ends with a human review note. It's designed for loop engineering workflows to automate draft generation of release notes.
Reads input from changelog-scan and applies a predefined structure and rules to categorize and format it. Generates a clean RELEASE_NOTES_DRAFT.md or a clear section for persistence. Highlights breaking changes and security issues at the beginning. Adds a human review note at the end. Only proposes a draft; it does not write or modify anything directly.
- Project maintainers who want to quickly convert a changelog into draft release notes before a release.
- Developers running loop engineering who want to automate release notes drafting as part of their weekly loop.
- Teams using Grok, Claude Code, or Codex who want consistent formatting for release notes drafts.
- Anyone who needs a draft release notes but wants to keep human review and approval in the loop.
How do you install this skill?
- Skill depends on external unverified Grok skill; core function may be unusable offline or from mainland China networks.
- Lacks installation and run instructions, making independent reproduction hard; suggest adding detailed documentation and examples.
- Data flow and sensitive information handling not explicit; review output content before use to avoid leakage.
- Local filesystem
Copy the skill folder (starters/changelog-drafter/.claude/skills/draft-release-notes/) into your Agent Skills directory (e.g., ~/.claude/skills/ for Claude Code or .claude/skills/ in your project). Alternatively, install the entire loop-engineering collection by running npx @cobusgreyling/loop-init . --pattern changelog-drafter. See the repository documentation for exact installation steps.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/cobusgreyling/loop-engineering.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/starters/changelog-drafter/.claude/skills/draft-release-notes" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
Once installed, send your agent any of these to trigger it:
- Draft release notes from this changelog: [paste changelog]
Provide the changelog output from changelog-scan as input. The skill will produce a structured RELEASE_NOTES_DRAFT.md file or a suggested section. Always review and edit the draft before applying. Example prompt: "Draft release notes from this changelog: [paste changelog]". Ensure changelog-scan has been run in the same loop.
What are this skill's strengths and limitations?
- Saves time by automating formatting.
- Ensures consistent structure in release notes.
- Flags breaking changes and security issues, reducing risk of omission.
- Designed as a propose-only skill, reducing accidental changes.
- Integrates seamlessly with loop engineering workflows.
- Requires changelog-scan output as input; not a standalone skill.
- Format may not perfectly match custom release workflows.
- Only generates a draft; human review and editing are required.
- No built-in tests or validation.
- Only as good as the changelog input; may not handle non-standard changelog formats well.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Release Notes Drafting Skill this page | 32 · Not recommended | ★ 11k | 3d ago | MIT |
| DeepChat Release Assistant | 46 · Use with care | ★ 6.4k | 3d ago | Apache-2.0 |
| Universal Release Flow | 51 · Use with care | ★ 4.3k | 18d ago | MIT |
| Git Workflow & Release Discipline | 54 · Use with care | ★ 103k | 8d ago | MIT |
| Cognitive-Load-Friendly Doc Design | 43 · Not recommended | ★ 7.6k | 3d ago | MIT |
How did FollowSkills review this skill?
Evidence shows the skill proposes drafts only, no auto-publish, and workflow marks human review, minimal permissions. But lacks explicit handling of sensitive data, rollback, and dependency security; publisher unverified. Deduction for incomplete data-flow transparency and rollback.
Skill only instructs to follow exact structure from external Grok skill, no local implementation or scripts; depends on unverified external service, not reproducible. Missing error handling and failure feedback. Deduction for inability to verify key path from files, external dependency.
Clear scenario (turn changelog-scan output into release notes draft), but no stated boundaries, non-fit cases, or trigger conditions; environment fit lacks Chinese support and mainland reachability. Deduction for unclear boundaries and triggers.
Clear name and short description, MIT license, but missing install notes, maintenance responsibility, versioning, examples, FAQ; detailed dependency on external Grok skill absent. Deduction for incomplete doc layering, troubleshooting and governance.
Goal clear, but no verifiable output examples; completeness depends on external skill, direct usability not confirmed. Deduction for lack of output verification and comparative baseline, static assessment cannot confirm marginal value.
Skill itself has no tests or execution evidence; depends on external Grok skill, repo has CI but not covering this skill's key paths. Deduction for no third-party execution evidence, static read cannot independently reproduce.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →