divine

Étoiles11

Forks0

Mis à jour11 juillet 2026 à 21:55

Thorough, research-grounded code review of a change, PR, or MR — multi-lens (correctness, security, simplification, tests), severity-ranked with impact + fix, configurable depth, optional PR comments. Use when asked to review code, "do a code review", "review this PR/MR", "review my changes/diff/branch", or audit a changeset before merge.

Installation

Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.

Source

Alexander-Tyagunov

Alexander-Tyagunov/magician

Ouvrir le dépôt GitHub Voir les dépôts du créateur

Téléchargement

Métiers associésSOC

Basé sur la classification professionnelle SOC

Analystes en assurance qualité des logiciels et testeursProfessions informatiques et mathématiques·SOC 15-1253

SKILL.md

readonly

name	divine
description	Thorough, research-grounded code review of a change, PR, or MR — multi-lens (correctness, security, simplification, tests), severity-ranked with impact + fix, configurable depth, optional PR comments. Use when asked to review code, "do a code review", "review this PR/MR", "review my changes/diff/branch", or audit a changeset before merge.

$divine — Codex Adapter

Read ../../references/codex-adapter.md, then read ../../../skills/divine/SKILL.md and follow the source skill through that Codex adapter. Keep the source skill's gates (the depth-selection gate and the PR-posting confirmation gate), the context contract for dispatched reviewers, the adversarial-verification pass, and the completion criteria.

Codex equivalents:

Change context — use Codex's shell for gh/glab and git diff; read the PR/MR description and linked tickets for intent.
Depth gate — ask the user how deep via Codex's question/approval UI instead of AskUserQuestion.
Lenses — when agent tools are available, run correctness/security/simplification/tests as bounded agents with generic self-contained prompts (Goal, Scope, diff/intent inputs, constraints, evidence and Return format). Do not request Claude profile names. Otherwise perform the lenses locally.
Grounding — invoke $magic (or available official docs/web tooling) when the change needs external/domain evidence.
Posting — writing a review to a PR/MR is a side effect: confirm explicitly and use the correct account before any gh/glab write.
Implement fixes (optional) — only on explicit user confirmation, spin a Codex task to fix Critical/High findings; committing/pushing is a side effect requiring approval, the correct account, and a feature branch.
Monitor mode — only create a recurring automation when the user explicitly asks for scheduling. Otherwise poll a running forge CLI via an exec session. Keep it idempotent by head SHA and review-only; never implement or push.
Review artifacts — keep intermediate lens reports in a temporary directory unless the user explicitly requests durable repository artifacts. A review must not dirty the worktree.

Plus depuis ce dépôt

même dépôt

accelerate

Alexander-Tyagunov/magician

Systematic performance profiling and optimization with a mandatory baseline-first gate — measure before changing, re-measure after. Use when something is slow or you need to hit a latency/throughput target.

2026-07-1312

autopsy

Alexander-Tyagunov/magician

Blameless post-mortem / RCA — gathers facts, reconstructs a timeline, runs 5-Whys, defines action items, writes the post-mortem file and commits it. Use after an incident or outage.

2026-07-1312

conjure

Alexander-Tyagunov/magician

Structured design dialogue with a visual companion — produces an approved spec and design artifacts before any implementation begins. Use at the start of a feature, before writing code.

2026-07-1312

divine

Alexander-Tyagunov/magician

2026-07-1312

magic

Alexander-Tyagunov/magician

Use when the user asks to research, investigate, analyze, find out, explore, examine, audit, or evaluate something — structured multi-source research with consulting, library-doc search, web search, and guided output delivery.

2026-07-1312

scrutinize

Alexander-Tyagunov/magician

Multi-agent code review AND remediation — dispatches correctness, security, and simplification reviewers in parallel, consolidates findings, then fixes criticals/highs. Use when reviewing a diff or PR before shipping.

2026-07-1312

name	divine
description	Thorough, research-grounded code review of a change, PR, or MR — multi-lens (correctness, security, simplification, tests), severity-ranked with impact + fix, configurable depth, optional PR comments. Use when asked to review code, "do a code review", "review this PR/MR", "review my changes/diff/branch", or audit a changeset before merge.

$divine — Codex Adapter

Codex equivalents:

Change context — use Codex's shell for gh/glab and git diff; read the PR/MR description and linked tickets for intent.
Depth gate — ask the user how deep via Codex's question/approval UI instead of AskUserQuestion.
Lenses — when agent tools are available, run correctness/security/simplification/tests as bounded agents with generic self-contained prompts (Goal, Scope, diff/intent inputs, constraints, evidence and Return format). Do not request Claude profile names. Otherwise perform the lenses locally.
Grounding — invoke $magic (or available official docs/web tooling) when the change needs external/domain evidence.
Posting — writing a review to a PR/MR is a side effect: confirm explicitly and use the correct account before any gh/glab write.
Implement fixes (optional) — only on explicit user confirmation, spin a Codex task to fix Critical/High findings; committing/pushing is a side effect requiring approval, the correct account, and a feature branch.
Monitor mode — only create a recurring automation when the user explicitly asks for scheduling. Otherwise poll a running forge CLI via an exec session. Keep it idempotent by head SHA and review-only; never implement or push.
Review artifacts — keep intermediate lens reports in a temporary directory unless the user explicitly requests durable repository artifacts. A review must not dirty the worktree.