[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7881SDKX58RNEYP1CTXM6KQ":3,"skill-same:01M7881SDKX58RNEYP1CTXM6KQ":47},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M7881SDKX58RNEYP1CTXM6KQ","verify-app",null,"Verify a Chimaera change end-to-end against the real daemon, UI, and PTY layer before claiming it works. Use after changing daemon\u002Fserver\u002FPTY\u002FUI behavior — especially terminal, reconnect, resize, previews, or agent-launch flows, which have all had bugs that only reproduce live. Skip for pure docs\u002Fcomment changes.","martinappberg\u002Fchimaera","martinappberg","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F17358377?v=4","01M787ZFA9CVKJKRZWAMAQVA9N","Martin Kjellberg","01M787ZFB4YKBEAFBABXX26F53",7,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fverify-app\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: verify-app\ndescription: Verify a Chimaera change end-to-end against the real daemon, UI, and PTY layer before claiming it works. Use after changing daemon\u002Fserver\u002FPTY\u002FUI behavior — especially terminal, reconnect, resize, previews, or agent-launch flows, which have all had bugs that only reproduce live. Skip for pure docs\u002Fcomment changes.\n---\n\n# Verifying a Chimaera change\n\nChimaera's verification culture is explicit: **features are driven live before\nthey land, not just unit-tested.** Terminal state, reconnect semantics, resize,\nand agent integrations have repeatedly had bugs that pass tests but break against\nthe real thing. Your PR should say what you ran and what you observed.\n\n## 1. Tests + lints first (necessary, not sufficient)\n\n```sh\nnpm --prefix web-ui run check      # svelte-check, if you touched the UI\nnpm --prefix web-ui run test       # targeted reducer\u002Funit coverage\nnpm --prefix web-ui run build      # production bundle and generated assets\njust check                        # plugin assets, then fmt\u002Fclippy\u002Ftest in both Rust workspaces\n```\n\nBuild `web-ui\u002Fdist` before workspace Rust checks on a fresh checkout; the server\nembeds it. For a daemon-only change, an existing bundle satisfies that prerequisite.\n\n`crates\u002Fchimaera-pty` has real PTY tests (`tests.rs`, `snapshot.rs`) — extend\nthem when you change terminal state or snapshotting.\n\n## 2. Select scenarios from the changed contracts\n\nDo not pick scenarios from memory. Read the diff, then find the contracts it\ntouches: the area `AGENTS.md` map for each changed path, the\n[feature page](..\u002F..\u002F..\u002Fdocs\u002Ffeatures\u002FREADME.md) for the capability, and each\n`.claude\u002Frules\u002F*.md` bullet whose `paths` match. Every contract you find\nselects at least one scenario from section 4. If the diff touches a contract that\nno scenario covers, write one. Name the new scenario in the evidence.\n\n## 3. Drive the real flow\n\nBring up the dev loop (see the **develop** skill), then exercise the *actual*\npath you changed. Pick the first observation tool your session has.\n\n**a. The agent browser, when you run inside Chimaera.** A session that Chimaera\nlaunched has the chimaera MCP server's `browser_*` tools (in Claude Code:\n`mcp__chimaera__browser_*`). Use them for UI checks instead of ad-hoc headless\nChrome or Playwright scripts ([agent browser](..\u002F..\u002F..\u002Fdocs\u002Ffeatures\u002Fagent-browser.md)):\n\n1. `browser_navigate {url}` opens the page in your own off-screen page. Only\n   loopback, this host and nodes of the user's running jobs load. Read the\n   result's first line: it names the engine (`app` or `chromium`).\n2. `browser_snapshot` reads the page as text with `[eN]` refs. It is the cheap\n   default; take one after each step.\n3. `browser_click {ref}`, `browser_type {text, ref, submit}`, `browser_press\n   {key}` and `browser_select {ref, value}` drive it with real input. Use\n   `browser_wait` for text, a selector or network idle before you read again.\n4. `browser_console` and `browser_network` give the errors, failed requests\n   and slow requests since your last call. Run both before you claim a pass.\n5. `browser_screenshot` (a `ref` for one element) is the proof for the PR.\n   `browser_inspect` gives computed styles and the box; `browser_eval` runs a\n   script for what nothing else answers.\n6. To show the user the result, call `browser_show`. Else call `browser_close`.\n\nKnow the limits before you trust a result. A native `\u003Cselect>` takes\n`browser_select`, never a click. There is no right-click, no HTML5\ndrag-and-drop and no file upload. A `· throttled` result means the app window\nis minimized or hidden, so timers run slowly. The app engine loads pages\nthrough the hosting daemon's proxy, which never forwards an absolute `\u002Fapi\u002Fv1`\nor `\u002Fws\u002F` request. So Chimaera's own workbench (the isolated preview) is\nexpected to break on the app engine (found by reading the code, not yet seen\nlive). If the snapshot shows the workbench cannot connect, fall back to (b) or\n(c) for the workbench itself. The agent browser still fits the web apps the\nchange serves, such as a proxied app or a preview's page.\n\n**b. The browser-pane tooling** (a desktop session with preview tools):\n`read_page` for structure\u002Fcontent, `read_console_messages` + `preview_logs`\nfor errors, `read_network_requests` for API\u002FWS traffic, a screenshot for proof.\n\n**c. Neither.** Use the develop skill's manual isolated launcher, daemon\nterminal output, and the available browser or native-app inspection tools.\n\nReport any visual or network surface your session could not inspect.\n\n**In a Claude cloud session** (`CLAUDE_CODE_REMOTE=true`) there is no preview\ntooling: run the daemon headless and drive it over HTTP\u002FWS instead, starting from\n`scripts\u002Fsmoke-daemon.mjs`, and name in the PR what you could not check (see the\n[cloud sessions guide](..\u002F..\u002F..\u002Fdocs\u002Fagent-guides\u002Fcloud-sessions.md)).\n\n## 4. Scenarios\n\nMatch the exercise to what you touched:\n\n- **Terminal \u002F PTY** — spawn a session, type, scroll. Then **kill the socket and\n  reattach**: the exact screen must come back (scrollback, colors, cursor, title),\n  no lost bytes. Reattach from a second tab at a **different window size**.\n- **Resize \u002F resync** — change the divider drag or font size on the initiator:\n  it must reflow natively with **no flicker or scroll reset**. A *foreign* client\n  repaints after a coalescing window. Read the\n  [architecture guide](..\u002F..\u002F..\u002Fdocs\u002Fagent-guides\u002Farchitecture.md) (\"Resize repaint\n  refinement\") for the invariants before trusting your eyes.\n  (Known accepted gap: a resync while on the alternate screen can't restore the\n  primary screen's scrollback.)\n- **Previews** — open the file types you touched (image, markdown, csv\u002Ftsv incl.\n  gzip, pdf, sandboxed html). Confirm that ranged\u002Fpaged formats stay incremental,\n  whole-file formats enforce their size caps, and the client renders; watch\n  daemon and browser memory on a big file.\n- **Agent launch \u002F links** — launch an agent runtime, confirm attention badges\n  (needs-attention \u002F finished \u002F errored) derive from protocol events in chat\n  and Claude hooks in its TUI, and that a linked terminal grant is scoped and\n  audited in scrollback.\n- **Agent browser** — in the macOS app, have an agent navigate, snapshot, click\n  and type in a local web app and read its console. Share a pane by its chain\n  link, act in it, and confirm that the agent's next act tool says \"Paused\".\n  `browser_show` must not reload the page. With no app window on the\n  workspace, the Chromium engine answers or refuses with the fix;\n  `just browser-smoke` drives it against a real browser.\n- **Reconnect** — reload and reattach from another machine\u002Ftab. Chat sockets\n  (`\u002Fws\u002Fchat\u002F{id}`) replay the seq-numbered gap; the session-list bus\n  (`\u002Fws\u002Fevents`) sends the current full roster; PTY sockets rebuild the screen\n  from an escape-sequence snapshot.\n\n## 5. Resource discipline is part of \"works\"\n\nThe daemon must stay light on shared hosts; cluster workspaces normally run\ninside allocations. While exercising, sanity-check it stays within budget (~150 MB RSS target, \u003C1 core steady-state, no runaway buffers) —\na preview of a huge Parquet\u002FHTML file must never balloon memory. A change that\nworks but leaks or busy-loops is not done.\n\n## 6. Report with structured evidence\n\nRecord one entry per scenario, in the PR body:\n\n```\nScenario:  \u003Cname, and the contract that selected it>\nSteps:     \u003Cthe exact actions, in order>\nObserved:  \u003Cwhat happened, including numbers such as RSS or window size>\nEvidence:  \u003Cartifact: log excerpt, screenshot path, request id, command output>\nResult:    pass | fail | blocked (\u003Cwhat was missing>)\n```\n\nUse `blocked` when a check could not run: a tool is unavailable, credentials are\nmissing, or there is no HPC allocation. Do not report a blocked check as a\nfailure. Do not report it as a pass. Name what it would take to unblock it.\n\nExample: \"spawned a claude session, killed the ws, reattached from a 2nd tab at\n80×24 → identical screen; `just check` green.\" That sentence, with its evidence,\nis the deliverable reviewers look for.\n","# Verifying a Chimaera change\n\nChimaera's verification culture is explicit: **features are driven live before\nthey land, not just unit-tested.** Terminal state, reconnect semantics, resize,\nand agent integrations have repeatedly had bugs that pass tests but break against\nthe real thing. Your PR should say what you ran and what you observed.\n\n## 1. Tests + lints first (necessary, not sufficient)\n\n```sh\nnpm --prefix web-ui run check      # svelte-check, if you touched the UI\nnpm --prefix web-ui run test       # targeted reducer\u002Funit coverage\nnpm --prefix web-ui run build      # production bundle and generated assets\njust check                        # plugin assets, then fmt\u002Fclippy\u002Ftest in both Rust workspaces\n```\n\nBuild `web-ui\u002Fdist` before workspace Rust checks on a fresh checkout; the server\nembeds it. For a daemon-only change, an existing bundle satisfies that prerequisite.\n\n`crates\u002Fchimaera-pty` has real PTY tests (`tests.rs`, `snapshot.rs`) — extend\nthem when you change terminal state or snapshotting.\n\n## 2. Select scenarios from the changed contracts\n\nDo not pick scenarios from memory. Read the diff, then find the contracts it\ntouches: the area `AGENTS.md` map for each changed path, the\n[feature page](..\u002F..\u002F..\u002Fdocs\u002Ffeatures\u002FREADME.md) for the capability, and each\n`.claude\u002Frules\u002F*.md` bullet whose `paths` match. Every contract you find\nselects at least one scenario from section 4. If the diff touches a contract that\nno scenario covers, write one. Name the new scenario in the evidence.\n\n## 3. Drive the real flow\n\nBring up the dev loop (see the **develop** skill), then exercise the *actual*\npath you changed. Pick the first observation tool your session has.\n\n**a. The agent browser, when you run inside Chimaera.** A session that Chimaera\nlaunched has the chimaera MCP server's `browser_*` tools (in Claude Code:\n`mcp__chimaera__browser_*`). Use them for UI checks instead of ad-hoc headless\nChrome or Playwright scripts ([agent browser](..\u002F..\u002F..\u002Fdocs\u002Ffeatures\u002Fagent-browser.md)):\n\n1. `browser_navigate {url}` opens the page in your own off-screen page. Only\n   loopback, this host and nodes of the user's running jobs load. Read the\n   result's first line: it names the engine (`app` or `chromium`).\n2. `browser_snapshot` reads the page as text with `[eN]` refs. It is the cheap\n   default; take one after each step.\n3. `browser_click {ref}`, `browser_type {text, ref, submit}`, `browser_press\n   {key}` and `browser_select {ref, value}` drive it with real input. Use\n   `browser_wait` for text, a selector or network idle before you read again.\n4. `browser_console` and `browser_network` give the errors, failed requests\n   and slow requests since your last call. Run both before you claim a pass.\n5. `browser_screenshot` (a `ref` for one element) is the proof for the PR.\n   `browser_inspect` gives computed styles and the box; `browser_eval` runs a\n   script for what nothing else answers.\n6. To show the user the result, call `browser_show`. Else call `browser_close`.\n\nKnow the limits before you trust a result. A native `\u003Cselect>` takes\n`browser_select`, never a click. There is no right-click, no HTML5\ndrag-and-drop and no file upload. A `· throttled` result means the app window\nis minimized or hidden, so timers run slowly. The app engine loads pages\nthrough the hosting daemon's proxy, which never forwards an absolute `\u002Fapi\u002Fv1`\nor `\u002Fws\u002F` request. So Chimaera's own workbench (the isolated preview) is\nexpected to break on the app engine (found by reading the code, not yet seen\nlive). If the snapshot shows the workbench cannot connect, fall back to (b) or\n(c) for the workbench itself. The agent browser still fits the web apps the\nchange serves, such as a proxied app or a preview's page.\n\n**b. The browser-pane tooling** (a desktop session with preview tools):\n`read_page` for structure\u002Fcontent, `read_console_messages` + `preview_logs`\nfor errors, `read_network_requests` for API\u002FWS traffic, a screenshot for proof.\n\n**c. Neither.** Use the develop skill's manual isolated launcher, daemon\nterminal output, and the available browser or native-app inspection tools.\n\nReport any visual or network surface your session could not inspect.\n\n**In a Claude cloud session** (`CLAUDE_CODE_REMOTE=true`) there is no preview\ntooling: run the daemon headless and drive it over HTTP\u002FWS instead, starting from\n`scripts\u002Fsmoke-daemon.mjs`, and name in the PR what you could not check (see the\n[cloud sessions guide](..\u002F..\u002F..\u002Fdocs\u002Fagent-guides\u002Fcloud-sessions.md)).\n\n## 4. Scenarios\n\nMatch the exercise to what you touched:\n\n- **Terminal \u002F PTY** — spawn a session, type, scroll. Then **kill the socket and\n  reattach**: the exact screen must come back (scrollback, colors, cursor, title),\n  no lost bytes. Reattach from a second tab at a **different window size**.\n- **Resize \u002F resync** — change the divider drag or font size on the initiator:\n  it must reflow natively with **no flicker or scroll reset**. A *foreign* client\n  repaints after a coalescing window. Read the\n  [architecture guide](..\u002F..\u002F..\u002Fdocs\u002Fagent-guides\u002Farchitecture.md) (\"Resize repaint\n  refinement\") for the invariants before trusting your eyes.\n  (Known accepted gap: a resync while on the alternate screen can't restore the\n  primary screen's scrollback.)\n- **Previews** — open the file types you touched (image, markdown, csv\u002Ftsv incl.\n  gzip, pdf, sandboxed html). Confirm that ranged\u002Fpaged formats stay incremental,\n  whole-file formats enforce their size caps, and the client renders; watch\n  daemon and browser memory on a big file.\n- **Agent launch \u002F links** — launch an agent runtime, confirm attention badges\n  (needs-attention \u002F finished \u002F errored) derive from protocol events in chat\n  and Claude hooks in its TUI, and that a linked terminal grant is scoped and\n  audited in scrollback.\n- **Agent browser** — in the macOS app, have an agent navigate, snapshot, click\n  and type in a local web app and read its console. Share a pane by its chain\n  link, act in it, and confirm that the agent's next act tool says \"Paused\".\n  `browser_show` must not reload the page. With no app window on the\n  workspace, the Chromium engine answers or refuses with the fix;\n  `just browser-smoke` drives it against a real browser.\n- **Reconnect** — reload and reattach from another machine\u002Ftab. Chat sockets\n  (`\u002Fws\u002Fchat\u002F{id}`) replay the seq-numbered gap; the session-list bus\n  (`\u002Fws\u002Fevents`) sends the current full roster; PTY sockets rebuild the screen\n  from an escape-sequence snapshot.\n\n## 5. Resource discipline is part of \"works\"\n\nThe daemon must stay light on shared hosts; cluster workspaces normally run\ninside allocations. While exercising, sanity-check it stays within budget (~150 MB RSS target, \u003C1 core steady-state, no runaway buffers) —\na preview of a huge Parquet\u002FHTML file must never balloon memory. A change that\nworks but leaks or busy-loops is not done.\n\n## 6. Report with structured evidence\n\nRecord one entry per scenario, in the PR body:\n\n```\nScenario:  \u003Cname, and the contract that selected it>\nSteps:     \u003Cthe exact actions, in order>\nObserved:  \u003Cwhat happened, including numbers such as RSS or window size>\nEvidence:  \u003Cartifact: log excerpt, screenshot path, request id, command output>\nResult:    pass | fail | blocked (\u003Cwhat was missing>)\n```\n\nUse `blocked` when a check could not run: a tool is unavailable, credentials are\nmissing, or there is no HPC allocation. Do not report a blocked check as a\nfailure. Do not report it as a pass. Name what it would take to unblock it.\n\nExample: \"spawned a claude session, killed the ws, reattached from a 2nd tab at\n80×24 → identical screen; `just check` green.\" That sentence, with its evidence,\nis the deliverable reviewers look for.\n",{"name":5,"description":7},"main","018562a00d38ea6f3865b4faafdc462a09900b19",".claude\u002Fskills\u002Fverify-app",1,"https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera","https:\u002F\u002Fchimaera.sh\u002F","https:\u002F\u002Fgithub.com\u002Fmartinappberg","Genetics PhD student @ Stanford","2026-11-12 07:28:58",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add martinappberg\u002Fchimaera --skill verify-app",{"tool":39,"command":40},"bunx","bunx skills add martinappberg\u002Fchimaera --skill verify-app",{"tool":42,"command":43},"pnpm","pnpm dlx skills add martinappberg\u002Fchimaera --skill verify-app","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fverify-app\u002FSKILL.md",true,[],[48,55,62,69,76,83],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":19},"01M7881YD8EBQZBNS5PCQA6R02","worktree-lifecycle","Keep Chimaera's git worktrees and their Rust build output from filling the disk — when to remove your worktree, how to run scripts\u002Fworktree-gc (dry run, --apply, --self), what the SessionStart\u002FSessionEnd hooks do, and the hard rule never to touch an ACTIVE worktree. Use when your PR has merged, when disk space is low or a session-start hook flagged it, before a big cargo build (just check, cargo test --workspace, an app build), or when asked to clean up worktrees or target\u002F dirs.","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fworktree-lifecycle\u002FSKILL.md",[],[],{"skillNo":56,"name":57,"nameZh":6,"description":58,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":59,"categories":60,"keywords":61,"publishedAt":19},"01M7881NE4W5TX0GW5HYR17AK1","ship-pr","Open a pull request for Chimaera correctly — the CI gates that must pass, the Conventional-Commit prefix that drives the automatic version bump, and the [skip release] marker for docs\u002Fchore PRs that shouldn't ship a version. Use when creating a PR, choosing a commit\u002FPR title, or deciding whether a change should cut a release.","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fship-pr\u002FSKILL.md",[],[],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":66,"categories":67,"keywords":68,"publishedAt":19},"01M7881EXTJZZCG3HJ7RTQKGT9","refresh-public-messaging","Refresh Chimaera's public face — the README and the GitHub Pages site (site\u002Findex.html + site\u002Fdocs.html) — so it matches what the product actually does now, and brainstorm positioning with the maintainer. Use when the README or website has drifted behind shipped features, when revisiting the selling points \u002F tagline, or on a periodic \"is our public face still current and well-said?\" pass. Positioning is the maintainer's call — never rewrite it unilaterally.","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Frefresh-public-messaging\u002FSKILL.md",[],[],{"skillNo":70,"name":71,"nameZh":6,"description":72,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":73,"categories":74,"keywords":75,"publishedAt":19},"01M7881A8MT7ARFW595KXVCZTW","document-feature","Add or update a Chimaera feature page under docs\u002Ffeatures\u002F so the feature catalog stays true as capabilities ship. Use when a change adds or changes user-facing behavior — a new pane surface, route, agent capability, CLI command — or when you find a feature page that's stale or missing. Defines the page structure, the derived-vs-intent split, and the rule that a feat: PR must carry its doc update.","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdocument-feature\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},"01M788141DR3AFJH2JAVZJXFZG","develop","Run Chimaera locally and iterate on it — the daemon + Vite web-UI dev loop, ports, dev-mode auth, the build-embeds-UI gotcha, the isolated per-worktree daemon (chimaerad-isolated), and where things live. Use when starting to develop or run the app, when a change needs the real daemon\u002FUI to see, when working in a git worktree or alongside another daemon, or when the dev server won't come up (port in use, Node version, blank UI).","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdevelop\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},"01M7880XTENRQMGQYZ1975FQW8","debug-live-app","Debug a running Chimaera daemon + web UI against a live isolated preview — read daemon logs, the browser console, and network traffic; reproduce a failure; and recognize the common failure modes (blank UI, 401, port in use, HMR vs rebuild). Use when the app misbehaves at runtime, a change doesn't show up, the UI won't connect, or you need to see what the daemon is actually doing.","https:\u002F\u002Fgithub.com\u002Fmartinappberg\u002Fchimaera\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fdebug-live-app\u002FSKILL.md",[],[]]