[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7F44TX793K2M5VFHS5A4RCG":3,"skill-same:01M7F44TX793K2M5VFHS5A4RCG":58},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":25,"commitSha":26,"skillDirectory":27,"forksCount":28,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":45},"01M7F44TX793K2M5VFHS5A4RCG","review-walkthrough",null,"Generates an interactive HTML walkthrough for reviewing code changes. Use only when explicitly called.","lza6\u002FArmor-piercing-system-prompts","lza6","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F128880206?v=4","01M7EB3K7WPX5T28EJ2NVHE6EG","Chinese-tingfeng","01M7EB3K97E1KQ16QWKJMFNGX8",11,"MIT","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Freview-walkthrough\u002FSKILL.md",[],[],"2026-11-16 10:05:34",false,"---\nname: review-walkthrough\ndescription: Generates an interactive HTML walkthrough for reviewing code changes. Use only when explicitly called.\ndisable-model-invocation: true\n---\n\n# Review Walkthrough\n\nGenerate a self-contained HTML walkthrough of the current branch, with the final\ndiff split into logical steps, explanations, and critical review findings. Choose\nthe grouping and reading order for comprehension; commit order need not dictate\neither. Invoke as `\u002Freview-walkthrough:review-walkthrough` in Claude Code or\n`$review-walkthrough` in Codex.\n\n## Capture the branch diff\n\nRespect a user-specified base. Otherwise, use the current open PR's base when PR\nmetadata is available. Without a PR, discover the repository's default branch\nfrom `git symbolic-ref --quiet --short refs\u002Fremotes\u002Forigin\u002FHEAD` or the configured\nremote's equivalent. Do not assume it is `main`. If the selected base is missing\nor ambiguous, ask for it rather than silently choosing another branch.\n\nResolve the base and `HEAD` to commit IDs with\n`git rev-parse --verify --end-of-options \"${ref}^{commit}\"`, then compute their\nmerge base with `git merge-base`. Capture the complete patch using\n`git diff --no-ext-diff --no-textconv --no-color --src-prefix=a\u002F --dst-prefix=b\u002F \"$merge_base\" \"$head_sha\" --`.\nInitialize variables and run the commands that consume them in the same shell\ninvocation; shell variables do not persist between tool calls. Save the patch to\na temporary file outside the checkout and retain its actual path. Stop on a Git\nerror or an empty patch and explain the result.\n\nThis scope includes committed changes through the captured `HEAD`. Mention any\nuncommitted work that is excluded. Do not change branches, reset the checkout,\nor alter source files to prepare the review. Keep generated data and HTML outside\nthe checkout unless the user requests a particular output location; generated\nreview artifacts never belong in the captured patch.\n\nFor a GitHub PR, collect `owner`, `repo`, `pr_number`, and `head_sha` from its\nmetadata. Include them only when the PR head matches the captured `HEAD` and the\nchosen base matches the PR base. Otherwise use `null` and explain why comment\nexport is unavailable. A failed authentication or network request is not evidence\nthat no PR exists; report the failure and ask for the base if it cannot be\ndetermined. GitHub access is optional when the user supplies the base or confirms\nthere is no PR.\n\n## Shape the review\n\nRead the changed files and the surrounding source, tests, configuration, and\ndependency metadata needed to judge them. Keep the review read-only; do not run\nthe project's tests or install its dependencies merely to build the walkthrough.\n\nGroup related files into steps covering one concept each. Put definitions before\ntheir consumers, wiring after the components it connects, and tests after or\nbeside their subjects. Copy each complete per-file diff from the captured patch\nverbatim into exactly one step. Reordering files between steps is allowed;\nsplitting, rewriting, omitting, or inventing their patches is not. Preserve\nrename, binary, and mode-change entries even when they have no text hunks.\nBinary summaries identify changed files; their payloads are outside this review.\n\nWrite one explanation per step, usually 50–150 words, covering the change's\npurpose, design choices, trade-offs, and connections to other steps. Refer to\nactual identifiers. Write one review list per step covering bugs, security,\nvalidation, API design, or performance issues supported by the code. Report\nfindings with severity rather than suppressing minor issues. Use an empty list\nfor a step with no findings.\n\nExplanations and review bodies are HTML fragments. Use `\u003Cp>`, `\u003Cstrong>`, `\u003Cem>`,\n`\u003Ccode>`, `\u003Cpre>`, and `\u003Cul>`\u002F`\u003Cli>` for structure. Write literal `\u003C` and `>` in\nprose or snippets as `&lt;` and `&gt;`. Backticks and Markdown fences are literal\ntext here. Supported inline tags also include `\u003Csup>`, `\u003Csub>`, `\u003Ckbd>`, `\u003Cdel>`,\nand `\u003Ca href=\"https:\u002F\u002F…\">`. The page sanitizes fragments and converts them to\nMarkdown when preparing a PR comment. Only safe link targets survive; arbitrary\nattributes, scripts, and images do not. Tables export as text rows with cell\nseparators; captions and row order are preserved. Prefer prose, lists, and code\nblocks when their structure is sufficient.\n\nAnchor findings to a file in the same step and a line in that file's displayed\ndiff. Use `side: \"RIGHT\"` for additions or new-file context and `side: \"LEFT\"`\nfor deletions or old-file context. For a range, `line` and `end_line` use that\nsame side within one hunk. A finding spanning multiple steps belongs with the\nfile it addresses, or remains unanchored. An unanchored finding is still\ndisplayed but cannot become an inline PR comment.\n\n## Render the artifact\n\nUse a file-writing tool to create a JSON object with the fields below. Preserve\nthe literal patch as JSON string data, with no shell expansion.\n\n| Field | Value |\n|-------|-------|\n| `title` | Feature or review title |\n| `steps` | Array of objects with `sha` (step ID such as `step-1`), `message` (step title), `files` (paths in patch order), and `diff` (complete per-file patches) |\n| `explanations` | HTML strings, one per step |\n| `reviews` | Arrays of findings, one per step; each finding has `severity` (`high`, `medium`, or `low`), `title`, and HTML `body` |\n| `pr_meta` | Object with `owner`, `repo`, `pr_number`, and `head_sha`, or `null` |\n\nAn anchored finding also has `file`, `line`, and `side`, plus optional `end_line`.\nThe three arrays have equal lengths and corresponding entries. File paths match\nthe diff headers without their `a\u002F` or `b\u002F` prefix; a renamed file uses its new\npath. Set `side` explicitly so lines that occur on both sides are unambiguous.\n\nRun the bundled renderer with the actual paths created for this review:\n\n```bash\nuv run --no-project {baseDir}\u002Fscripts\u002Frender_walkthrough.py \\\n  --input \u002Ftmp\u002Fdata.json \\\n  --diff \u002Ftmp\u002Fcaptured.patch \\\n  --output \u002Ftmp\u002Fwalkthrough.html\n```\n\nThe renderer validates the data, compares every step's patches with the complete\ncaptured diff, checks anchors, and safely embeds the result in the bundled\ntemplate. Fix reported input errors rather than bypassing the renderer or\ngenerating an alternative page.\n\nOpen the output with `open` on macOS or `xdg-open` on Linux when a browser is\navailable, and tell the user its path. In a headless run, report the file without\nopening it. When PR metadata is present, the page lets the user copy a `gh api`\ncommand for their selected comments. It does not execute the command or post\nanything to GitHub. Comment and Request Changes require a review summary before\nthe command can be copied; the summary is optional for Approve.\n","# Review Walkthrough\n\nGenerate a self-contained HTML walkthrough of the current branch, with the final\ndiff split into logical steps, explanations, and critical review findings. Choose\nthe grouping and reading order for comprehension; commit order need not dictate\neither. Invoke as `\u002Freview-walkthrough:review-walkthrough` in Claude Code or\n`$review-walkthrough` in Codex.\n\n## Capture the branch diff\n\nRespect a user-specified base. Otherwise, use the current open PR's base when PR\nmetadata is available. Without a PR, discover the repository's default branch\nfrom `git symbolic-ref --quiet --short refs\u002Fremotes\u002Forigin\u002FHEAD` or the configured\nremote's equivalent. Do not assume it is `main`. If the selected base is missing\nor ambiguous, ask for it rather than silently choosing another branch.\n\nResolve the base and `HEAD` to commit IDs with\n`git rev-parse --verify --end-of-options \"${ref}^{commit}\"`, then compute their\nmerge base with `git merge-base`. Capture the complete patch using\n`git diff --no-ext-diff --no-textconv --no-color --src-prefix=a\u002F --dst-prefix=b\u002F \"$merge_base\" \"$head_sha\" --`.\nInitialize variables and run the commands that consume them in the same shell\ninvocation; shell variables do not persist between tool calls. Save the patch to\na temporary file outside the checkout and retain its actual path. Stop on a Git\nerror or an empty patch and explain the result.\n\nThis scope includes committed changes through the captured `HEAD`. Mention any\nuncommitted work that is excluded. Do not change branches, reset the checkout,\nor alter source files to prepare the review. Keep generated data and HTML outside\nthe checkout unless the user requests a particular output location; generated\nreview artifacts never belong in the captured patch.\n\nFor a GitHub PR, collect `owner`, `repo`, `pr_number`, and `head_sha` from its\nmetadata. Include them only when the PR head matches the captured `HEAD` and the\nchosen base matches the PR base. Otherwise use `null` and explain why comment\nexport is unavailable. A failed authentication or network request is not evidence\nthat no PR exists; report the failure and ask for the base if it cannot be\ndetermined. GitHub access is optional when the user supplies the base or confirms\nthere is no PR.\n\n## Shape the review\n\nRead the changed files and the surrounding source, tests, configuration, and\ndependency metadata needed to judge them. Keep the review read-only; do not run\nthe project's tests or install its dependencies merely to build the walkthrough.\n\nGroup related files into steps covering one concept each. Put definitions before\ntheir consumers, wiring after the components it connects, and tests after or\nbeside their subjects. Copy each complete per-file diff from the captured patch\nverbatim into exactly one step. Reordering files between steps is allowed;\nsplitting, rewriting, omitting, or inventing their patches is not. Preserve\nrename, binary, and mode-change entries even when they have no text hunks.\nBinary summaries identify changed files; their payloads are outside this review.\n\nWrite one explanation per step, usually 50–150 words, covering the change's\npurpose, design choices, trade-offs, and connections to other steps. Refer to\nactual identifiers. Write one review list per step covering bugs, security,\nvalidation, API design, or performance issues supported by the code. Report\nfindings with severity rather than suppressing minor issues. Use an empty list\nfor a step with no findings.\n\nExplanations and review bodies are HTML fragments. Use `\u003Cp>`, `\u003Cstrong>`, `\u003Cem>`,\n`\u003Ccode>`, `\u003Cpre>`, and `\u003Cul>`\u002F`\u003Cli>` for structure. Write literal `\u003C` and `>` in\nprose or snippets as `&lt;` and `&gt;`. Backticks and Markdown fences are literal\ntext here. Supported inline tags also include `\u003Csup>`, `\u003Csub>`, `\u003Ckbd>`, `\u003Cdel>`,\nand `\u003Ca href=\"https:\u002F\u002F…\">`. The page sanitizes fragments and converts them to\nMarkdown when preparing a PR comment. Only safe link targets survive; arbitrary\nattributes, scripts, and images do not. Tables export as text rows with cell\nseparators; captions and row order are preserved. Prefer prose, lists, and code\nblocks when their structure is sufficient.\n\nAnchor findings to a file in the same step and a line in that file's displayed\ndiff. Use `side: \"RIGHT\"` for additions or new-file context and `side: \"LEFT\"`\nfor deletions or old-file context. For a range, `line` and `end_line` use that\nsame side within one hunk. A finding spanning multiple steps belongs with the\nfile it addresses, or remains unanchored. An unanchored finding is still\ndisplayed but cannot become an inline PR comment.\n\n## Render the artifact\n\nUse a file-writing tool to create a JSON object with the fields below. Preserve\nthe literal patch as JSON string data, with no shell expansion.\n\n| Field | Value |\n|-------|-------|\n| `title` | Feature or review title |\n| `steps` | Array of objects with `sha` (step ID such as `step-1`), `message` (step title), `files` (paths in patch order), and `diff` (complete per-file patches) |\n| `explanations` | HTML strings, one per step |\n| `reviews` | Arrays of findings, one per step; each finding has `severity` (`high`, `medium`, or `low`), `title`, and HTML `body` |\n| `pr_meta` | Object with `owner`, `repo`, `pr_number`, and `head_sha`, or `null` |\n\nAn anchored finding also has `file`, `line`, and `side`, plus optional `end_line`.\nThe three arrays have equal lengths and corresponding entries. File paths match\nthe diff headers without their `a\u002F` or `b\u002F` prefix; a renamed file uses its new\npath. Set `side` explicitly so lines that occur on both sides are unambiguous.\n\nRun the bundled renderer with the actual paths created for this review:\n\n```bash\nuv run --no-project {baseDir}\u002Fscripts\u002Frender_walkthrough.py \\\n  --input \u002Ftmp\u002Fdata.json \\\n  --diff \u002Ftmp\u002Fcaptured.patch \\\n  --output \u002Ftmp\u002Fwalkthrough.html\n```\n\nThe renderer validates the data, compares every step's patches with the complete\ncaptured diff, checks anchors, and safely embeds the result in the bundled\ntemplate. Fix reported input errors rather than bypassing the renderer or\ngenerating an alternative page.\n\nOpen the output with `open` on macOS or `xdg-open` on Linux when a browser is\navailable, and tell the user its path. In a headless run, report the file without\nopening it. When PR metadata is present, the page lets the user copy a `gh api`\ncommand for their selected comments. It does not execute the command or post\nanything to GitHub. Comment and Request Changes require a review summary before\nthe command can be copied; the summary is optional for Approve.\n",{"name":5,"description":7,"disable-model-invocation":24},true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","05-技能与代理开发\u002Freview-walkthrough",3,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts","https:\u002F\u002Fgithub.com\u002Flza6","个人vx：Tf00798\r\n个人✈：@TFGY999","2026-10-08 14:29:22",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":24},[35,38,41],{"tool":36,"command":37},"npx","npx skills add lza6\u002FArmor-piercing-system-prompts --skill review-walkthrough",{"tool":39,"command":40},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill review-walkthrough",{"tool":42,"command":43},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill review-walkthrough","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Freview-walkthrough\u002FSKILL.md",[46,52],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"agents\u002Fopenai.yaml","REFERENCE","参考文档",234,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Freview-walkthrough\u002Fagents\u002Fopenai.yaml",{"path":53,"resourceType":54,"resourceTypeDesc":55,"sizeBytes":56,"sourceHtmlUrl":57},"scripts\u002Fpyproject.toml","SCRIPT","脚本",110,"https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Freview-walkthrough\u002Fscripts\u002Fpyproject.toml",[59,66,71,78,85,92],{"skillNo":60,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":63,"categories":64,"keywords":65,"publishedAt":19},"01M7FFG485KY83TD6HSC4YPCMC","obsidian-clipper-template-creator","Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator\u002FSKILL.md",[],[],{"skillNo":67,"name":61,"nameZh":6,"description":62,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":19},"01M7FFG0T6AZ0STEJPSB3CX84K","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobsidian-clipper-template-creator-antigravity-awesome-skills-main\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":19},"01M7FFFY0QEM14SKWPDYGMXM2P","python-api","Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP\u002FmAP semantics, and source-style import diagnostics.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Fpython-api\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":19},"01M7FFFVZQZ2GJ0C0TGP7JHD20","file-evaluation","Folder-based PASCAL VOC AP\u002FmAP evaluation from ground-truth and detection text files.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002Fsub-skills\u002Ffile-evaluation\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":19},"01M7FFFSFNYSE77SJ6PE1CK2DA","object-detection-metrics","Guides legacy Object-Detection-Metrics PASCAL VOC AP\u002FmAP evaluation from detection text folders and direct Python API classes.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fobject-detection-metrics\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7FFFP0GCK1RVXBCPQ8APBMT","nx-workspace-patterns","Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F07-后端与架构\u002Fnx-workspace-patterns\u002FSKILL.md",[],[]]