SkillsAdd

dpolivaev/spec-loop

spec-loop-review-change

>- Review an existing change from a local or user-trusted source, such as a pull request, merge request, branch diff, commit range, or agent-written code. Use when the user needs a merge verdict, a file-wise review walk-through, reconstruction of the change after the fact, or both.

First seen Jul 23, 2026

Installation

$ npx skills add dpolivaev/spec-loop --skill spec-loop-review-change
View on skills.sh ↗ GitHub repository ↗

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

flutter-add-widget-preview Similar name
flutter/agent-plugins

Adds interactive widget previews to the project using the previews.dart system. Use when creati…

29.4K installs
recipe-review-overdue-tasks Similar name
googleworkspace/cli

Find Google Tasks that are past due and need attention.

27.2K installs
recipe-review-meet-participants Similar name
googleworkspace/cli

Review who attended a Google Meet conference and for how long.

26.9K installs
security-review Similar name
getsentry/skills

Security code review for vulnerabilities.

15.5K installs
web-design-reviewer Similar name
github/awesome-copilot

This skill enables visual inspection of websites running locally or remotely to identify and fi…

13.3K installs
sql-code-review Similar name
github/awesome-copilot

Universal SQL code review assistant that performs comprehensive security, maintainability, and …

13K installs

Also in this package

Other skills from dpolivaev/spec-loop · top by installs.

npx skills add dpolivaev/spec-loop
spec-loop-plan-task 58 spec-loop-write-glossary 55 spec-loop-implementation-flow 54 spec-loop-assess-pull-request 53 spec-loop-setup-doc-rendering 53 spec-loop-clarify-task 50 spec-loop-compact-task-file 47 spec-loop-write-adr 46 spec-loop-prepare-implementation-approval 45 spec-loop-prepare-execution-approval 11

Browse all from dpolivaev/spec-loop

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Repository health

Stars 34
License LICENSE.md
Default branch main
Open issues 0
Status Active

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,958 B
  • docs SUMMARY.md 310 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 2 installs

SKILL.md

Spec Loop Review Change

Use this optional skill for work that is already implemented, not for planning or executing new work.

If trust, the exact change to review, or the comparison range is unclear, stop and ask before fetching GitHub, GitLab, or Git data.

Read [review-core-guidance.md](./review-core-guidance.md) first.

Then choose the review mode and evidence source together, and load only the needed guidance files. Load both mode files only when the user wants both outputs:

  • Assessment — pick this when the user wants a verdict,

recommendation, PR or MR review file, or reconstruction of the change after the fact. Load [assessment-guidance.md](./assessment-guidance.md).

  • Walk-through — pick this when the user wants file ownership,

review blocks, reading order, or inline file-wise explanation. Load [walk-through-guidance.md](./walk-through-guidance.md).

If the review target is a specific GitHub or GitLab pull request or merge request page, number, or other remote review target, also load [github-gitlab-evidence-guidance.md](./github-gitlab-evidence-guidance.md). Do not load github-gitlab-evidence-guidance.md for branch diffs, commit ranges, current local changes, or other local Git review targets.

Do not load spec-loop-plan-task by default. Only the assessment mode reuses parts of it, and only for the specific conventions it actually needs.

Use [spec-loop-clarify-task](../spec-loop-clarify-task/SKILL.md) when an unresolved choice could materially change trust, evidence source, comparison range, review mode, document layout, or verdict expectations.