[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-84037":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":10,"languages":10,"totalLinesOfCode":10,"stars":11,"forks":12,"watchers":13,"openIssues":14,"contributorsCount":14,"subscribersCount":14,"size":14,"stars1d":14,"stars7d":15,"stars30d":15,"stars90d":14,"forks30d":14,"starsTrendScore":15,"compositeScore":16,"rankGlobal":10,"rankLanguage":10,"license":10,"archived":17,"fork":17,"defaultBranch":18,"hasWiki":19,"hasPages":17,"topics":20,"createdAt":10,"pushedAt":10,"updatedAt":21,"readmeContent":22,"aiSummary":10,"trendingCount":14,"starSnapshotCount":14,"syncStatus":23,"lastSyncTime":24,"discoverSource":25},84037,"review-forge","vikingmute\u002Freview-forge","vikingmute","review-forge is an Agent Skill for structured, auditable code review workflows","",null,99,8,98,0,1,2.86,false,"main",true,[],"2026-06-12 02:04:37","# Review Forge\n\n[![skills.sh](https:\u002F\u002Fskills.sh\u002Fb\u002Fvikingmute\u002Freview-forge)](https:\u002F\u002Fskills.sh\u002Fvikingmute\u002Freview-forge)\n\n`review-forge` is an Agent Skill for structured, auditable code review\nworkflows. It orchestrates multi-pass LLM review, finding synthesis,\nhuman-approved fix selection, implementation, regression testing, independent\nverification, and status tracking.\n\n## What It Does\n\n- Reviews local changes, branches, PRs, or commit ranges.\n- Synthesizes multiple review perspectives into one fix checklist.\n- Uses checkboxes as the human approval boundary for fixes.\n- Fixes only approved items.\n- Requires tests after fixes unless testing is blocked and documented.\n- Verifies fixes independently when possible.\n- Keeps process artifacts isolated under `code_review\u002F`.\n\n## Commands\n\n- `review`: create one model-specific review file.\n- `synthesize`: merge review files in one feature folder into `summary.md`.\n- `fix`: fix checked items, run tests, and update status.\n- `verify`: verify fixed items, inspect or rerun tests, and update status.\n\n## Default Review Scope\n\nWhen no target or base is specified:\n\n1. If there are uncommitted or staged changes, Review Forge reviews the working\n   tree diff.\n2. If the working tree is clean, it reviews `main...HEAD`.\n3. If `main` does not exist, it tries `master...HEAD`.\n4. If no reasonable base can be inferred, it asks for one.\n\nExplicit PRs, branches, commit ranges, or base refs always override these\ndefaults.\n\n## Artifacts\n\nReview Forge groups workflow files by feature under `code_review\u002F`:\n\n```text\ncode_review\u002F\u003Cfeature>\u002F\n  codex.md\n  opencode.md\n  cursor.md\n  summary.md\n  fix-plan.md\n  verify.md\n  status.md\n```\n\nEach model review should produce one file named after the model or agent. If\nthe same model runs multiple perspectives, use a suffix such as\n`codex-security.md`.\n\nCommand outputs are intentionally simple:\n\n- `review` creates one model review file only, for example `codex.md`.\n- `synthesize` creates `summary.md` only.\n- `fix` creates or updates `fix-plan.md` and `status.md`.\n- `verify` creates or updates `verify.md` and `status.md`.\n\nTimestamped files such as `STATUS_20260529.md` are legacy-style outputs and\nshould not be created unless you explicitly ask to preserve historical runs.\n\nAdd `code_review\u002F` to the target repository's `.gitignore` unless you\nintentionally want to commit review process files.\n\n## Language Policy\n\nThe skill instructions, template field names, and machine-readable status enums\nare in English. Generated reports can use `report_language: auto` to follow the\nuser's prompt language, or explicit values such as `en`, `zh-CN`, or `ja`.\n\nStatus values remain stable English enums, while display labels may be localized.\n\n## Installation\n\nInstall from GitHub with the Skills CLI:\n\n```sh\nnpx skills add vikingmute\u002Freview-forge\n```\n\nOr use the full GitHub URL:\n\n```sh\nnpx skills add https:\u002F\u002Fgithub.com\u002Fvikingmute\u002Freview-forge\n```\n\nInstall globally for a specific agent:\n\n```sh\nnpx skills add vikingmute\u002Freview-forge -g -a codex\n```\n\nInstall globally for multiple agents:\n\n```sh\nnpx skills add vikingmute\u002Freview-forge -g -a codex -a cursor -a opencode\n```\n\nUpdate an installed copy:\n\n```sh\nnpx skills update review-forge\n```\n\nUpdate only global installs:\n\n```sh\nnpx skills update review-forge -g\n```\n\nFor local development without publishing, clone it into the cross-client Agent\nSkills directory:\n\n```sh\ngit clone https:\u002F\u002Fgithub.com\u002Fvikingmute\u002Freview-forge.git ~\u002F.agents\u002Fskills\u002Freview-forge\n```\n\nFor a project-local install, copy or clone it into the project:\n\n```sh\nmkdir -p .agents\u002Fskills\ngit clone https:\u002F\u002Fgithub.com\u002Fvikingmute\u002Freview-forge.git .agents\u002Fskills\u002Freview-forge\n```\n\nSome clients also support their own native skill locations. For Codex, this is\nalso valid:\n\n```sh\ncp -R review-forge ~\u002F.codex\u002Fskills\u002Freview-forge\n```\n\nThen invoke it from a compatible agent with:\n\n```text\nUse $review-forge to review this branch.\n```\n\nIf a client does not support explicit `$skill` syntax, ask it to use the\n`review-forge` skill or select it from the client's skills UI.\n\n## Example Prompts\n\n```text\nUse $review-forge review feature: checkout-refactor model: codex.\n```\n\n```text\nUse $review-forge review feature: checkout-refactor model: opencode perspective: security.\n```\n\n```text\nUse $review-forge synthesize feature: checkout-refactor.\n```\n\n```text\nUse $review-forge fix feature: checkout-refactor.\n```\n\n```text\nUse $review-forge verify feature: checkout-refactor.\n```\n",2,"2026-06-11 04:12:08","CREATED_QUERY"]