[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7EXP63DF78VYCJXBJV7TCW5":3,"skill-same:01M7EXP63DF78VYCJXBJV7TCW5":46},{"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},"01M7EXP63DF78VYCJXBJV7TCW5","cate-cli",null,"Drive Cate browser, terminal, editor, panel, review, and coding-agent orchestration surfaces from a Cate terminal. Browser page automation targets Cate's live webviews directly.","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\u002Fcate-cli\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: cate-cli\ndescription: Drive Cate browser, terminal, editor, panel, review, and coding-agent orchestration surfaces from a Cate terminal. Browser page automation targets Cate's live webviews directly.\nuser-invocable: true\n---\n\n# Cate CLI\n\n`cate` is available inside Cate terminals and agent shells. It talks to the\ncurrent workspace and requires the relevant Settings → CLI permission.\n\nStart by listing panels:\n\n```bash\ncate panel list\n```\n\nWhen working repeatedly with one panel, select it for the current agent or\nterminal session:\n\n```bash\ncate panel set 1a2b3c4d\ncate panel current\n```\n\nThe selection is isolated by a per-terminal CLI session, so other agents and\nterminals keep their own targets. Short ids from `panel list`\nare accepted. Use `--panel \u003Cid>` only as a one-command override. Clear the\nselection to return to Cate's automatic focused\u002Fgrouped resolution:\n\n```bash\ncate panel clear\n```\n\nSelections can point to any native panel. Browser and terminal commands reject\na selected panel of the wrong type instead of silently controlling another\npanel. If a selected panel was closed, select another panel before continuing.\n\n## Browser workflow\n\nBrowser control uses persistent JavaScript with the `cua` tab API. The old argv\nactions, selectors, page evaluation, and revisioned string refs have been removed.\nStart by binding a tab to get its accessibility state, then request a screenshot\nwhen visual context is useful:\n\n```bash\ncate browser run 'var tab = await cua.getTab({panelId:\"\u003Cfull-panel-id>\"});'\ncate browser run 'await tab.getAXStateAndScreenshot();'\n```\n\nUse full panel IDs inside JavaScript. `--panel \u003Cid>` supports short IDs as an\noverride for CLI panel resolution. Discover tabs with `await cua.listTabs()`.\nCreate a tab with `await cua.createBrowserTab(\"https:\u002F\u002Fexample.com\")`, or pass\n`{panelId:tab.panelId}` as the second argument to choose its panel. Pass\n`{newPanel:true}` to create a separate panel; Cate also creates one when needed.\nBindings pin both panel and tab; they never silently follow a user's tab switch.\n\nUse numeric IDs from the latest AX observation. For example, after observing a form\ncontaining textbox 17 and button 42:\n\n```bash\ncate browser run 'await tab.setValue(17,\"user@example.com\"); await tab.click(42); await tab.waitFor({url:\"**\u002Fdashboard\"});'\ncate browser run 'await tab.getAXStateAndScreenshot();'\n```\n\nDo not guess IDs or coordinates. Observations contain `kind`, `observationId`,\n`documentId`, URL, title and viewport. AX observations (`kind:\"ax\"`) also contain\naccessibility state and structured elements with role, name, value and states.\n`getAXState()` normally emits a concise diff; `getAXState({disableDiffing:true})`\nemits the full tree. `getScreenshot()` returns only viewport pixels\u002Fidentity\n(`kind:\"image\"`, empty state\u002Felements), avoiding an AX scan. It does not refresh\nnumeric IDs. `getAXStateAndScreenshot()` refreshes both together. The SDK keeps\nthe last AX observation for numeric targets and the latest visual observation\nfor coordinates. `{emit:false}` suppresses automatic output. `{profile:true}`\nadds phase timings, image bytes and estimated retained-cache usage. Each code\ncell retains at most 16 million serialized observation characters, including\n`emit:false`; split long screenshot loops across cells.\n\nThe SDK carries the latest observation through each action and emits fresh state.\nNumeric IDs persist within one document; navigation requires fresh IDs. Coordinate\nactions use `[x,y]` in observed viewport CSS pixels and reject stale viewport\ncoordinates. A dispatched click is not proof that a business operation completed:\nuse `waitFor` or inspect the resulting state.\n\n```javascript\nawait tab.click(42);                         \u002F\u002F or [x,y]\nawait tab.setValue(17, \"replacement\");\nawait tab.typeText(\"insert at selection\");\nawait tab.pressKey(\"Return\");\nawait tab.selectText(17, \"text\", {selectionType:\"cursor_after\"});\nawait tab.scroll([400,300], \"down\", 1);\nawait tab.drag([100,100], [300,200]);\nawait tab.setChecked(42, true);\nawait tab.selectOption(42, [\"DE\"]);\nawait tab.upload(42, \"\u002Fauthorized\u002Ffile\");\nawait tab.waitFor({text:\"Saved\"});\nawait tab.waitFor({element:42, state:\"enabled\"});\nawait tab.goto(\"https:\u002F\u002Fexample.com\");\nawait tab.back(); await tab.forward(); await tab.reload();\nawait tab.setViewport({width:1280,height:800});\nawait tab.resize({width:800,height:600});\nawait tab.downloads(); await tab.close();\n```\n\nKeep deterministic batches short and inspect unexpected changes before continuing.\nUse `var` for reusable bindings; top-level `await` is supported. The session has\nno Node.js, filesystem, network, or DOM evaluation access. Await every action.\n`nodeRepl.write(value)` adds text output. `cate browser reset` clears JavaScript\nbindings without closing tabs. Reset and timeout cancel queued and pending browser\nactions; input already dispatched cannot be undone. Sessions are isolated per\nterminal\u002Fagent through `CATE_CLI_SESSION_ID`; timed-out sessions reset.\n`typeText` resolves current focus before inserting at the current selection,\nincluding fields inside frames and shadow roots.\n\n`cate browser run 'await tab.getAXStateAndScreenshot();'` returns AX state and\nsaves a screenshot to a temporary PNG file. Open the printed path with your image-viewing tool before visual\nreasoning. Shell output cannot itself attach pixels to the model. `--json`\nreturns structured content with base64 image data; base64 text is not visual\ninput. The same screenshot output works from `tab.getScreenshot()` in code.\nAX reads remain available in code for deterministic branches and extraction.\n\nAgent actions display a cursor and click ripples in the browser panel, without\nfield bounding-box highlights. Filling and typing animate the cursor at the\nedited field. User input takes control back and cancels pending automation. Responsive viewport size and canvas\npanel size are independent; `resize` applies only to canvas panels with a 400×300\nminimum.\n\n## Other surfaces\n\n```bash\ncate editor open src\u002Fapp.tsx:42\ncate panel create terminal\ncate panel create canvas\ncate panel set \u003Cid>\ncate panel current\ncate panel clear\ncate panel close \u003Cid>\n```\n\nRead a terminal before sending input. `type` does not append Enter:\n\n```bash\ncate panel set 1a2b3c4d\ncate terminal read\ncate terminal type npm test\ncate terminal press enter\n```\n\nTerminal input goes to whatever currently owns that PTY, including foreground\nTUIs. Never send keys until the panel id and current screen are verified.\n\n## Agent orchestration\n\nUse `cate agent` when a task benefits from visible, persistent delegation:\nindependent parallel work, cross-provider review, or isolated implementation in\na Cate worktree. Keep small, tightly coupled edits in the current agent.\n\nDiscover registered runs before acting on an older mission or after context\ncompaction:\n\n```bash\ncate agent list\n```\n\nCreate a worker with a bounded, self-contained prompt and concrete success\ncriteria. Cate chooses the first hook-ready registered agent when `--agent` is\nomitted:\n\n```bash\ncate agent create \"Inspect the API boundary and report risks\" --title \"API scout\"\ncate agent create \"Implement the parser and run its focused tests\" \\\n  --agent codex --title \"Parser\" --new-worktree agent\u002Fparser\ncate agent create \"Review the current worktree changes\" --worktree \u003Cworktree-id>\ncate agent create \"Continue in this terminal\" --terminal \u003Cterminal-panel-id>\n```\n\n`--terminal` reuses an idle terminal panel by restarting its PTY with the\nselected agent. It rejects busy terminals, agent-owned terminals, and the\ncalling terminal. Omit it to create a new terminal panel.\n\nWorkers may recursively create and supervise their own workers with the same\ncommands. This naturally forms an agent tree: each terminal owns the workers it\ncreates, and each parent normally communicates with its direct children. Use\nrecursion when another level of decomposition is genuinely useful, not merely\nto relay a simple instruction.\n\nSupervise workers through the agent lifecycle rather than typing into their\nterminals:\n\n```bash\ncate agent wait \u003Crun-id> [\u003Crun-id>...] --wait-timeout 10000\ncate agent inspect \u003Crun-id>\ncate agent send \u003Crun-id> \"Please add the missing regression test\"\ncate agent review \u003Crun-id>\ncate agent apply \u003Crun-id>\ncate agent keep \u003Crun-id>\ncate agent discard \u003Crun-id>\ncate agent stop \u003Crun-id>\n```\n\nRun ids may be the unique short ids printed by `cate agent list`. `wait` accepts\n5000–60000 milliseconds and may be called with no ids to monitor all live\ndirect children. Call it again while workers remain active. `inspect` includes\nrecent terminal output; use `cate terminal read --panel \u003Cpanel-id>` only as a\nlower-level diagnostic fallback.\n\nPrefer `send` for follow-up work on the same responsibility. If\n`followUpSupported` is false, create a fresh worker instead. When a worker fails,\ninspect `failureReason`; a provider-specific authentication, quota, or service\nfailure can justify retrying with a different registered `--agent`.\n\nFor an isolated worker, ask it to run relevant checks and commit completed work,\nthen use `review` before choosing `apply`, `keep`, or `discard`. Apply rechecks\nthat the worktree is clean and mergeable. Discard permanently removes a\nworker-owned worktree and its branch, including uncommitted changes, without an\ninteractive confirmation. Keep records that the worktree should remain for\nlater. Review is read-only: a finished process or successful review does not\nmean its branch has been integrated. The parent remains responsible for\nverification and for reporting any uncommitted or unintegrated work.\n\n## Review panels\n\nSelect a Review Panel once, inspect its comparison, and record structured\nfindings. `--panel \u003Cid>` is an optional one-command override for every review\ncommand.\n\n```bash\ncate panel set \u003Creview-panel-id>\ncate review inspect\ncate review note add --file src\u002Fapp.ts --line 42 --side new \\\n  --severity error --body \"Handle the rejected request\"\ncate review note resolve \u003Cnote-id>\ncate review complete\n```\n\nUse `complete` only when running as the review agent assigned by that Review\nPanel. Review commands record findings; they do not modify files, stage,\ncommit, or push changes.\n","# Cate CLI\n\n`cate` is available inside Cate terminals and agent shells. It talks to the\ncurrent workspace and requires the relevant Settings → CLI permission.\n\nStart by listing panels:\n\n```bash\ncate panel list\n```\n\nWhen working repeatedly with one panel, select it for the current agent or\nterminal session:\n\n```bash\ncate panel set 1a2b3c4d\ncate panel current\n```\n\nThe selection is isolated by a per-terminal CLI session, so other agents and\nterminals keep their own targets. Short ids from `panel list`\nare accepted. Use `--panel \u003Cid>` only as a one-command override. Clear the\nselection to return to Cate's automatic focused\u002Fgrouped resolution:\n\n```bash\ncate panel clear\n```\n\nSelections can point to any native panel. Browser and terminal commands reject\na selected panel of the wrong type instead of silently controlling another\npanel. If a selected panel was closed, select another panel before continuing.\n\n## Browser workflow\n\nBrowser control uses persistent JavaScript with the `cua` tab API. The old argv\nactions, selectors, page evaluation, and revisioned string refs have been removed.\nStart by binding a tab to get its accessibility state, then request a screenshot\nwhen visual context is useful:\n\n```bash\ncate browser run 'var tab = await cua.getTab({panelId:\"\u003Cfull-panel-id>\"});'\ncate browser run 'await tab.getAXStateAndScreenshot();'\n```\n\nUse full panel IDs inside JavaScript. `--panel \u003Cid>` supports short IDs as an\noverride for CLI panel resolution. Discover tabs with `await cua.listTabs()`.\nCreate a tab with `await cua.createBrowserTab(\"https:\u002F\u002Fexample.com\")`, or pass\n`{panelId:tab.panelId}` as the second argument to choose its panel. Pass\n`{newPanel:true}` to create a separate panel; Cate also creates one when needed.\nBindings pin both panel and tab; they never silently follow a user's tab switch.\n\nUse numeric IDs from the latest AX observation. For example, after observing a form\ncontaining textbox 17 and button 42:\n\n```bash\ncate browser run 'await tab.setValue(17,\"user@example.com\"); await tab.click(42); await tab.waitFor({url:\"**\u002Fdashboard\"});'\ncate browser run 'await tab.getAXStateAndScreenshot();'\n```\n\nDo not guess IDs or coordinates. Observations contain `kind`, `observationId`,\n`documentId`, URL, title and viewport. AX observations (`kind:\"ax\"`) also contain\naccessibility state and structured elements with role, name, value and states.\n`getAXState()` normally emits a concise diff; `getAXState({disableDiffing:true})`\nemits the full tree. `getScreenshot()` returns only viewport pixels\u002Fidentity\n(`kind:\"image\"`, empty state\u002Felements), avoiding an AX scan. It does not refresh\nnumeric IDs. `getAXStateAndScreenshot()` refreshes both together. The SDK keeps\nthe last AX observation for numeric targets and the latest visual observation\nfor coordinates. `{emit:false}` suppresses automatic output. `{profile:true}`\nadds phase timings, image bytes and estimated retained-cache usage. Each code\ncell retains at most 16 million serialized observation characters, including\n`emit:false`; split long screenshot loops across cells.\n\nThe SDK carries the latest observation through each action and emits fresh state.\nNumeric IDs persist within one document; navigation requires fresh IDs. Coordinate\nactions use `[x,y]` in observed viewport CSS pixels and reject stale viewport\ncoordinates. A dispatched click is not proof that a business operation completed:\nuse `waitFor` or inspect the resulting state.\n\n```javascript\nawait tab.click(42);                         \u002F\u002F or [x,y]\nawait tab.setValue(17, \"replacement\");\nawait tab.typeText(\"insert at selection\");\nawait tab.pressKey(\"Return\");\nawait tab.selectText(17, \"text\", {selectionType:\"cursor_after\"});\nawait tab.scroll([400,300], \"down\", 1);\nawait tab.drag([100,100], [300,200]);\nawait tab.setChecked(42, true);\nawait tab.selectOption(42, [\"DE\"]);\nawait tab.upload(42, \"\u002Fauthorized\u002Ffile\");\nawait tab.waitFor({text:\"Saved\"});\nawait tab.waitFor({element:42, state:\"enabled\"});\nawait tab.goto(\"https:\u002F\u002Fexample.com\");\nawait tab.back(); await tab.forward(); await tab.reload();\nawait tab.setViewport({width:1280,height:800});\nawait tab.resize({width:800,height:600});\nawait tab.downloads(); await tab.close();\n```\n\nKeep deterministic batches short and inspect unexpected changes before continuing.\nUse `var` for reusable bindings; top-level `await` is supported. The session has\nno Node.js, filesystem, network, or DOM evaluation access. Await every action.\n`nodeRepl.write(value)` adds text output. `cate browser reset` clears JavaScript\nbindings without closing tabs. Reset and timeout cancel queued and pending browser\nactions; input already dispatched cannot be undone. Sessions are isolated per\nterminal\u002Fagent through `CATE_CLI_SESSION_ID`; timed-out sessions reset.\n`typeText` resolves current focus before inserting at the current selection,\nincluding fields inside frames and shadow roots.\n\n`cate browser run 'await tab.getAXStateAndScreenshot();'` returns AX state and\nsaves a screenshot to a temporary PNG file. Open the printed path with your image-viewing tool before visual\nreasoning. Shell output cannot itself attach pixels to the model. `--json`\nreturns structured content with base64 image data; base64 text is not visual\ninput. The same screenshot output works from `tab.getScreenshot()` in code.\nAX reads remain available in code for deterministic branches and extraction.\n\nAgent actions display a cursor and click ripples in the browser panel, without\nfield bounding-box highlights. Filling and typing animate the cursor at the\nedited field. User input takes control back and cancels pending automation. Responsive viewport size and canvas\npanel size are independent; `resize` applies only to canvas panels with a 400×300\nminimum.\n\n## Other surfaces\n\n```bash\ncate editor open src\u002Fapp.tsx:42\ncate panel create terminal\ncate panel create canvas\ncate panel set \u003Cid>\ncate panel current\ncate panel clear\ncate panel close \u003Cid>\n```\n\nRead a terminal before sending input. `type` does not append Enter:\n\n```bash\ncate panel set 1a2b3c4d\ncate terminal read\ncate terminal type npm test\ncate terminal press enter\n```\n\nTerminal input goes to whatever currently owns that PTY, including foreground\nTUIs. Never send keys until the panel id and current screen are verified.\n\n## Agent orchestration\n\nUse `cate agent` when a task benefits from visible, persistent delegation:\nindependent parallel work, cross-provider review, or isolated implementation in\na Cate worktree. Keep small, tightly coupled edits in the current agent.\n\nDiscover registered runs before acting on an older mission or after context\ncompaction:\n\n```bash\ncate agent list\n```\n\nCreate a worker with a bounded, self-contained prompt and concrete success\ncriteria. Cate chooses the first hook-ready registered agent when `--agent` is\nomitted:\n\n```bash\ncate agent create \"Inspect the API boundary and report risks\" --title \"API scout\"\ncate agent create \"Implement the parser and run its focused tests\" \\\n  --agent codex --title \"Parser\" --new-worktree agent\u002Fparser\ncate agent create \"Review the current worktree changes\" --worktree \u003Cworktree-id>\ncate agent create \"Continue in this terminal\" --terminal \u003Cterminal-panel-id>\n```\n\n`--terminal` reuses an idle terminal panel by restarting its PTY with the\nselected agent. It rejects busy terminals, agent-owned terminals, and the\ncalling terminal. Omit it to create a new terminal panel.\n\nWorkers may recursively create and supervise their own workers with the same\ncommands. This naturally forms an agent tree: each terminal owns the workers it\ncreates, and each parent normally communicates with its direct children. Use\nrecursion when another level of decomposition is genuinely useful, not merely\nto relay a simple instruction.\n\nSupervise workers through the agent lifecycle rather than typing into their\nterminals:\n\n```bash\ncate agent wait \u003Crun-id> [\u003Crun-id>...] --wait-timeout 10000\ncate agent inspect \u003Crun-id>\ncate agent send \u003Crun-id> \"Please add the missing regression test\"\ncate agent review \u003Crun-id>\ncate agent apply \u003Crun-id>\ncate agent keep \u003Crun-id>\ncate agent discard \u003Crun-id>\ncate agent stop \u003Crun-id>\n```\n\nRun ids may be the unique short ids printed by `cate agent list`. `wait` accepts\n5000–60000 milliseconds and may be called with no ids to monitor all live\ndirect children. Call it again while workers remain active. `inspect` includes\nrecent terminal output; use `cate terminal read --panel \u003Cpanel-id>` only as a\nlower-level diagnostic fallback.\n\nPrefer `send` for follow-up work on the same responsibility. If\n`followUpSupported` is false, create a fresh worker instead. When a worker fails,\ninspect `failureReason`; a provider-specific authentication, quota, or service\nfailure can justify retrying with a different registered `--agent`.\n\nFor an isolated worker, ask it to run relevant checks and commit completed work,\nthen use `review` before choosing `apply`, `keep`, or `discard`. Apply rechecks\nthat the worktree is clean and mergeable. Discard permanently removes a\nworker-owned worktree and its branch, including uncommitted changes, without an\ninteractive confirmation. Keep records that the worktree should remain for\nlater. Review is read-only: a finished process or successful review does not\nmean its branch has been integrated. The parent remains responsible for\nverification and for reporting any uncommitted or unintegrated work.\n\n## Review panels\n\nSelect a Review Panel once, inspect its comparison, and record structured\nfindings. `--panel \u003Cid>` is an optional one-command override for every review\ncommand.\n\n```bash\ncate panel set \u003Creview-panel-id>\ncate review inspect\ncate review note add --file src\u002Fapp.ts --line 42 --side new \\\n  --severity error --body \"Handle the rejected request\"\ncate review note resolve \u003Cnote-id>\ncate review complete\n```\n\nUse `complete` only when running as the review agent assigned by that Review\nPanel. Review commands record findings; they do not modify files, stage,\ncommit, or push changes.\n",{"name":5,"description":7,"user-invocable":24},true,"main","d69096e239f2a89aabc2cd078c47664a9d3b1e11","05-技能与代理开发\u002Fcate-cli",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 cate-cli",{"tool":39,"command":40},"bunx","bunx skills add lza6\u002FArmor-piercing-system-prompts --skill cate-cli",{"tool":42,"command":43},"pnpm","pnpm dlx skills add lza6\u002FArmor-piercing-system-prompts --skill cate-cli","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\u002Fcate-cli\u002FSKILL.md",[],[47,54,61,68,76,83],{"skillNo":48,"name":49,"nameZh":6,"description":50,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":51,"categories":52,"keywords":53,"publishedAt":19},"01M7EXRM12046G0GM1WGMMFXCN","ci-cd-and-automation","Automates CI\u002FCD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates, configure test runners in CI, or establish deployment strategies.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fci-cd-and-automation\u002FSKILL.md",[],[],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":19},"01M7EXRFYYSJ3Y15AGPDDEWWQD","chrome-devtools","Uses Chrome DevTools via MCP for efficient debugging, troubleshooting and browser automation. Use when debugging web pages, automating browser interactions, analyzing performance, or inspecting network requests. This skill does not apply to `--slim` mode (MCP configuration).","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fchrome-devtools\u002FSKILL.md",[],[],{"skillNo":62,"name":63,"nameZh":6,"description":64,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":65,"categories":66,"keywords":67,"publishedAt":19},"01M7EXR6EYPE4X5X1BTAE6QJRV","chrome-cdp","Drive a real Chrome via the Chrome DevTools Protocol (CDP). Ensures a\ndebuggable Chrome is running (launching it if needed) so the agent can\nnavigate, read page content, inspect the DOM, run JavaScript, and take\nscreenshots — without the user typing any launch command. Includes a\nfirst-class \"review a live running app\" flow: point Chrome at the app under\ndevelopment and inspect\u002Fscreenshot the REAL page (what a static-artifact tool\ncannot do). Backed by Glimpse and the chrome-devtools MCP server. Use when\nasked to \"use chrome\", \"open chrome\", \"drive my browser\", \"read the page in\nchrome\", \"review the running app\", \"screenshot the app\", or \"\u002Fchrome-cdp\".","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fchrome-cdp\u002FSKILL.md",[],[],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":72,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7EXR1Y7E79KQKG4J798XGH6","check-identity-pack","Run an AFP 100-point or AUSTRAC safe-harbour identity check over a set of documents, and report exactly what's missing. Use when the user asks to check identity documents, verify someone's ID for onboarding, or assess whether a document pack satisfies Australian identity requirements.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fcheck-identity-pack\u002FSKILL.md",[],[],{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":82,"publishedAt":19},"01M7EXQXW7C7PGREJW9ZZTZ16H","changelog","Auto-generates a changelog from git commits, sprint data, and design documents. Produces both internal and player-facing versions.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fchangelog\u002FSKILL.md",[],[],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":89,"publishedAt":19},"01M7EXQMH2ZME3ANN8Q8MEWPFB","cellxgene-census","Query the CZ CELLxGENE Census programmatically for versioned public single-cell and spatial transcriptomics data. Use when you need population-scale cell metadata, gene expression slices, Census summary counts, source H5AD URIs\u002Fdownloads, embeddings, spatial Census data, or reference atlas comparisons across organisms, tissues, diseases, assays, and cell types. For analyzing your own local single-cell data use scanpy, anndata, or scvi-tools.","https:\u002F\u002Fgithub.com\u002Flza6\u002FArmor-piercing-system-prompts\u002Fblob\u002Fmain\u002F05-技能与代理开发\u002Fcellxgene-census\u002FSKILL.md",[],[]]