Cross-reviews an existing change (MR/PR, branch or commit range) with two independent reviewers, a fresh subagent and a model from another family through its CLI, who then cross-examine each other's findings; you settle one ledger and fix the confirmed findings when asked. Use when the user asks for a cross review (交叉 review, 交叉审查) of a change.
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# Cross-review a change
You own the review of one change: pin it, start two reviewers, run their cross-exam, settle every finding and report. The two reviewers are a fresh subagent of your model and a model from another family through its CLI. You are the arbiter: where they disagree, you read the code at the head and decide. Before merging a large MR, when the review should also run the app and fix through sub-MRs until clean, [merge-gate](../merge-gate/SKILL.md) does that.
The reviewers work from merge-gate's [review brief](../merge-gate/references/review-brief.md): what to review, the fields of each finding, the severity scale, the cross-exam and the boundary. Pass on the user's own instructions about scope and focus as written.
## 1. Pin the change
1. Resolve full SHAs for the head and the base: an MR's diff refs from the platform's CLI or API, or the merge-base of a branch with its target.
2. The evidence directory holds the input, the reports and the ledger: the location the user names; otherwise `.cross-review/<id>/` in your worktree, excluded through `.git/info/exclude`.
…
The line at the top of this page pins the label git:20261024.56119b0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
GET https://markdownregistry.com/api/v1/artifacts/art_bsuntdyjyjbmz2ck
GET https://markdownregistry.com/api/v1/resolve?ref=coder-xieshijie/dev-skills/cross-review
GET https://markdownregistry.com/api/v1/blob/fa860c9ea6e35684f430258dc2e3a44ca7d531d206c32c2e51d1af07e871b154
Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
coder-xieshijie/dev-skills · skills/agent-prompt-rules/SKILL.md · Design rules and a change process for prompts written for agents, multi-agent pipelines and SKILL.md files, based on…
coder-xieshijie/dev-skills · skills/core-grill/SKILL.md · Question the user in rounds to settle the decisions that change user-visible results, record terms, and compile a…
coder-xieshijie/dev-skills · skills/core-spec/SKILL.md · Converges a conversation and its materials into spec.md, the core decisions; when delivery will be automated, also…
coder-xieshijie/dev-skills · skills/deliver/SKILL.md · One owner implements a requirement and delivers a mergeable MR/PR. Use after spec.md and verify.md are confirmed and…
coder-xieshijie/dev-skills · skills/dev-feedback/SKILL.md · Runs stage D after a requirement is delivered, or over a span of time. Gathers what the sessions, the MR and the…
coder-xieshijie/dev-skills · skills/explain-as-fool/SKILL.md · Explains a topic in plain, literal language for someone who knows nothing about it. Use when someone new to a topic…
coder-xieshijie/dev-skills · skills/merge-gate/SKILL.md · Gates a large MR/PR before it merges. Independent review and runtime verification of its latest head, fixes through…
coder-xieshijie/dev-skills · skills/mr-for-human/SKILL.md · Turns an MR, PR or code diff into a reading guide for people, with the important decisions first and then a reading…
coder-xieshijie/dev-skills · skills/plan-for-agents/SKILL.md · Creates, revises or checks a complete plan for agents to execute, turning requirements and confirmed decisions into an…
coder-xieshijie/dev-skills · skills/recon-to-contract/SKILL.md · Converges a comparison of two or more external references into one executable contract with evidence, decisions and…
smart-ai-memory/attune-ai · .claude/skills/cross-review/SKILL.md · One-shot second-opinion review of a real diff by a DIFFERENT model (Claude, Codex or Antigravity) — advisory only…
smart-ai-memory/attune-ai · plugin/skills/cross-review/SKILL.md · One-shot second-opinion review of a real diff by a DIFFERENT model (Claude, Codex or Antigravity) — advisory only…
smart-ai-memory/attune-ai · .agents/skills/cross-review/SKILL.md · One-shot second-opinion review of a real diff by a DIFFERENT model (Claude, Codex or Antigravity) — advisory only…
dmitriyyukhanov/agent-plugins · plugins/codex-collaboration/skills/cross-review/SKILL.md · Parallel dual review between Claude and Codex. Both models review independently. Findings confirmed by reviewer…
dmitriyyukhanov/claude-plugins · plugins/codex-collaboration/skills/cross-review/SKILL.md · Parallel dual review between Claude and Codex. Both models review independently. Findings confirmed by reviewer…
sentimony/skills · plugins/skills/skills/cross-review/SKILL.md · You MUST use this when an implementation plan or a finished implementation needs an independent review by other agent…
broomva/skills · skills/governance/cross-review/SKILL.md · bstack P20 — Cross-Model Adversarial Review Gate. The model that wrote the code cannot be the final judge of the code…
zzoretich/unidex-agent-harness · codex-plugin/skills/cross-review/SKILL.md · Harden a plan with an independent review by the other provider (Claude Code and Codex review each other), then…
