[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6AE9FS2X5X7WN6TXAY75ZR4":3,"skill-same:01M6AE9FS2X5X7WN6TXAY75ZR4":89},{"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":23,"publishedAt":50,"linkOnly":51,"markdownContent":52,"markdownBody":53,"frontMatter":54,"refName":65,"commitSha":66,"skillDirectory":67,"forksCount":68,"repositoryHtmlUrl":69,"repositoryHomepageUrl":70,"ownerHtmlUrl":71,"ownerBlogUrl":72,"ownerDescription":73,"updatedAt":74,"install":75,"resources":88},"01M6AE9FS2X5X7WN6TXAY75ZR4","curviate-quickstart",null,"The first run: install the Curviate CLI, authenticate this machine with `curviate setup`, confirm with `curviate doctor`, connect a LinkedIn account, and prove the whole path works by reading the same profile live and then from the store. Use when Curviate is being set up for the first time, when `curviate` is not installed or not authenticated, when a command fails with a credential or account error, or when someone asks how to get started. Reads only: it issues no write of any kind.","davila7\u002Fclaude-code-templates","davila7","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6216945?v=4","01M6ADEWWFPS191NFM0Y1EX1JH","Daniel Avila","01M6ADJHGM78PHQRD3XZTFDVKK",33332,"MIT","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fcurviate\u002Fcurviate-quickstart\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49],"quickstart","curviate quickstart","curviate","when","cli","reading","error","how to","get","time","up","only","install","write","command","how","Recruiting","SET","https","issues","doctor","run","connect","plugin","setup","github","2026-11-10 10:12:30",false,"---\nname: curviate-quickstart\ndescription: \"The first run: install the Curviate CLI, authenticate this machine with `curviate setup`, confirm with `curviate doctor`, connect a LinkedIn account, and prove the whole path works by reading the same profile live and then from the store. Use when Curviate is being set up for the first time, when `curviate` is not installed or not authenticated, when a command fails with a credential or account error, or when someone asks how to get started. Reads only: it issues no write of any kind.\"\nlicense: MIT\nmetadata:\n  version: 0.1.0\n  author: Curviate\n  tags: [LinkedIn, CLI, Agents, Sales, Recruiting, Outreach]\n  repository: https:\u002F\u002Fgithub.com\u002FCurviate\u002Fcurviate-plugin\n---\n\n# Curviate: the first run\n\nFour steps, in this order: install, authenticate, connect an account, prove it. Every command here\nis a read. Nothing in this skill posts, messages, invites, follows or endorses anyone.\n\nEstablished against CLI `0.33.0`.\n\n## 1. Install the CLI\n\n```bash\nnpm install -g @curviate\u002Fcli\ncurviate --version    # needs Node 18 or newer\n```\n\n## 2. Authenticate this machine: `curviate setup`\n\n`setup` opens the dashboard, takes the short code it shows you, and saves an API key into a local\nprofile. **The key is never printed and never passed on a command line.**\n\nThe command has two shapes, and which one you get is decided by whether **stdout** is a terminal.\n\n### Agent shape: two legs, because a browser has to happen in between\n\nWhen stdout is not a terminal, or when `--json` is passed, `setup` prints a JSON object and exits\n`0`. It does not prompt and it does not open a browser:\n\n```bash\ncurviate setup --json\n# {\"authorize_url\":\"…\",\"next_step\":\"curviate setup --code -\",\"instructions\":\"…\"}\n```\n\nOpen `authorize_url` in a browser signed in to the workspace, press Authorize, and read back the\ncode it displays. Then hand that code to the second leg **on stdin**:\n\n```bash\nprintf '%s' \"$CODE\" | curviate setup --code -\n# {\"ok\":true,\"tenant\":\"…\",\"profile\":\"default\",\"account_id\":\"…\"}\n```\n\n- **`--code -` reads from stdin deliberately.** Passing the code as a flag value works but warns:\n  a value on the command line is visible to other processes through `ps` and is saved in shell\n  history.\n- **The two legs must run on the same machine, under the same configuration directory.** Leg one\n  writes a short-lived resume file beside the config file, readable only by you, holding the session\n  id and the private key that opens the delivered credential. Leg two reads it and deletes it. Split\n  the legs across machines, users, containers or config-directory overrides and the exchange cannot\n  complete.\n- **`--code -` with no setup in progress exits `2`** and says so. Run leg one first.\n- The code is short, case-insensitive, and drops anything that is not one of its own characters, so\n  a pasted code with stray spaces or punctuation is still accepted.\n\n### Human shape\n\nWith stdout on a terminal and no `--json`, `setup` prints the link, opens a browser, and prompts for\nthe code (up to three attempts). `--no-browser` prints the link rather than opening it, which is what\nyou want over SSH or anywhere a browser would open on the wrong machine.\n\nOne asymmetry to know before scripting it: **stdout on a terminal with stdin piped is refused at exit\n`2`**, because there is nowhere to paste a code into. It names `CURVIATE_API_KEY` and\n`curviate login --api-key -` as the alternatives. A *fully* piped invocation is not that case: it is\nthe agent shape above, and it exits `0`.\n\n### What each failure means\n\n| Exit | Cause | What to do |\n|---|---|---|\n| `0` | Authenticated. The key is saved to the named profile. | Continue to `doctor`. |\n| `2` | Usage: no setup in progress, or a terminal was needed and there was none. | Fix the invocation. Never retry unchanged. |\n| `3` | The code was wrong, expired, or the attempts ran out; or the workspace has no API key to deliver; or the key saved but a verifying call did not succeed. **Read the message: these are deliberately four different sentences, and only the first is fixed by starting over.** | Wrong or expired code: run `setup` again for a fresh one. No API key: create one in the dashboard first. Saved but unverified: run `doctor`. |\n| `6` | Too many attempts against this endpoint. | Wait, then try again. |\n| `7` | The API could not be reached. | Check the network, and the base URL if you overrode it. |\n| `1` | The exchange failed in a way the CLI did not expect, or the delivered credential could not be opened by this process. | Worth one retry. If it repeats it is a bug to report, not a state to work around. |\n\nA code is short-lived and allows only a handful of attempts. If in doubt, start a fresh `setup`\nrather than reusing anything.\n\n## 3. Confirm: `curviate doctor`\n\n```bash\ncurviate doctor --json\n```\n\nIt reports the CLI version, the config path, the active profile, the base URL, where the credential\nresolved from, **which workspace it authenticated as**, whether the API is reachable, whether the\ncredential is accepted, and every connected account with its status.\n\nThree things that surprise people:\n\n- **Run `setup` first, then `doctor`, not the other way round.** Reachability is only tested through\n  a real call, so with no credential resolved `doctor` reports the API as unreachable. That is not a\n  network problem; there was nothing to call with.\n- **The workspace line is populated only for a credential `setup` wrote.** A key supplied through the\n  environment or `login` authenticates perfectly well, but `doctor` has no name to show for it and\n  says so. And where a name is shown it identifies the workspace; it is not promised to be a\n  friendly, human-chosen label.\n- **Zero connected accounts is a pass, not a failure.** `doctor` reports `0 connected` and exits `0`.\n  That is the expected state of a workspace nobody has connected an account to yet.\n\n`doctor` reports rather than throws, so it prints a report and not an error envelope. Branch on the\nexit code and read `checks[]` for which stage failed:\n\n- **`0`**: every check passed.\n- **`2`**: the request was refused before it was sent, so nothing reached the network. An empty\n  credential, or a malformed base URL (the report's detail reads `Invalid base URL: expected an\n  absolute http:\u002F\u002F or https:\u002F\u002F URL`). Fix the invocation or the value; a retry cannot help.\n- **`3`**: no credential resolved, or one resolved and was rejected. Run `setup`.\n- **`7`**: the API could not be reached, or was reached and answered with a platform fault. Both are\n  worth a retry.\n\n**This list is not exhaustive, and the codes beyond it are not `doctor`'s own.** The credential check\nis a real API call, and whatever refusal comes back is passed straight through, so any code the API\ncan produce can surface here. The one a first run meets most often is **exit `5` on a workspace with\nno active seat**: the credential is fine and the network is fine, and no amount of re-running `setup`\nwill change it. Look an unfamiliar code up in the area skill for the surface you are about to use:\neach ends with a table of the codes its commands produce and the action each implies.\n\nThe `checks[]` entries separate the three stages, and the wording is exact: `api reachable` reading\n`not checked` means the request never left this machine, while `could not reach` means it left and\nfound nothing. `credential valid` reading `not checked` means nothing asked the credential anything;\nit is not a verdict on the credential, so do not go looking for a bad key on the strength of it.\n\n## 4. Connect a LinkedIn account\n\nCurviate acts on LinkedIn through an account you connect. Until one is connected, authentication is\ncomplete and correct while **every account-scoped command is unusable**: that is the whole state,\nand it is not an error to be debugged.\n\n```bash\ncurviate account list --json    # empty on a fresh workspace\n```\n\nIf `setup` reported an `account_id`, an account is already connected **and already set as this\nprofile's default**; skip the rest of this step, including the block below, and go to step 5.\n\nOtherwise, find a free seat first:\n\n```bash\ncurviate account seats --json    # seat_id of any entry with \"occupied\": false\n```\n\nThen connect one with `curviate account link --seat-id \u003Cid> ...`. It attaches a LinkedIn account to\nthat seat and usually needs a verification code, which means a human. Non-interactive shells get exit\n`12` and finish through the checkpoint flow. **The command surface, its flags and the checkpoint\ncommands are in `curviate-profile`**: read that before running it, because a half-finished connect\nleaves a session to poll rather than a clean failure.\n\nThen tell the CLI which account to act as:\n\n```bash\ncurviate account list --json              # the acc_id of the account you just connected\ncurviate config set-account \u003Cacc_id>      # make it this profile's default\n```\n\n**Do not skip this.** `setup` stores a default account only when the workspace already had one to\nhand back, which is exactly what the expected first run does not have. Without a default, every\naccount-scoped command, including the demonstration below, stops at exit `2` asking for `--account`.\nSet it once here, or pass `--account \u003Cacc_id>` on each call. **`config set-account` writes to the\nstored profile, so it needs one**: if your credential came from `CURVIATE_API_KEY` rather than\n`setup`, it exits `2` saying there is no config file, and `--account \u003Cacc_id>` per call (or\n`CURVIATE_ACCOUNT` in the environment) is the path for you. Account ids change when an account is\nreconnected, so read one live rather than hard-coding it.\n\n## 5. Prove the whole path: the same read, twice\n\nThis is the demonstration, and it is worth running because it exercises authentication, the\nconnected account, projection and the store in one pass. Both calls are reads of your own profile.\n\n```bash\ncurviate profile me --mode live --json          # \"source\":\"live\"\ncurviate profile me --mode cache_only --json    # \"source\":\"store\", with \"observed_at\"\n```\n\nThe first forces a live call to LinkedIn. The second refuses to go to LinkedIn at all and answers\nfrom the copy the first one stored, so it comes back with `source: store` and an `observed_at`\ntimestamp saying when that copy was taken.\n\n**Read `source` rather than assuming it.** That field, not the freshness you asked for, is what\nactually happened. The full set of modes, and what a `cache_only` read does when nothing is stored,\nare in `curviate-profile`.\n\n## Full command surface\n\n\u003C!-- generated: command surface, CLI 0.33.0 -->\n\nRead from the CLI's own `--help` at version 0.33.0. Descriptions, traps and confidence\ntags elsewhere in this skill are hand-written and carry the version they were established against.\n\nEvery command below that takes flags at all also accepts `--base-url`, `--json`, `--profile`.\n\n| Command | Arguments | Flags |\n|---|---|---|\n| `curviate setup` | *(none)* | `--no-browser`, `--code` |\n| `curviate doctor` | *(none)* | `--api-key`, `--timeout` |\n| `curviate login` | *(none)* | `--api-key`, `--account` |\n\n\u003C!-- \u002Fgenerated -->\n\n## Where to go next\n\nBack to `curviate` for the dispatch table. The common first tasks: look someone up\n(`curviate-profile`), find a set of people (`curviate-search`), read your inbox (`curviate-inbox`).\nEach area skill ends with the exit codes its commands can produce and what each one implies; branch\non those, never on the message text.\n","# Curviate: the first run\n\nFour steps, in this order: install, authenticate, connect an account, prove it. Every command here\nis a read. Nothing in this skill posts, messages, invites, follows or endorses anyone.\n\nEstablished against CLI `0.33.0`.\n\n## 1. Install the CLI\n\n```bash\nnpm install -g @curviate\u002Fcli\ncurviate --version    # needs Node 18 or newer\n```\n\n## 2. Authenticate this machine: `curviate setup`\n\n`setup` opens the dashboard, takes the short code it shows you, and saves an API key into a local\nprofile. **The key is never printed and never passed on a command line.**\n\nThe command has two shapes, and which one you get is decided by whether **stdout** is a terminal.\n\n### Agent shape: two legs, because a browser has to happen in between\n\nWhen stdout is not a terminal, or when `--json` is passed, `setup` prints a JSON object and exits\n`0`. It does not prompt and it does not open a browser:\n\n```bash\ncurviate setup --json\n# {\"authorize_url\":\"…\",\"next_step\":\"curviate setup --code -\",\"instructions\":\"…\"}\n```\n\nOpen `authorize_url` in a browser signed in to the workspace, press Authorize, and read back the\ncode it displays. Then hand that code to the second leg **on stdin**:\n\n```bash\nprintf '%s' \"$CODE\" | curviate setup --code -\n# {\"ok\":true,\"tenant\":\"…\",\"profile\":\"default\",\"account_id\":\"…\"}\n```\n\n- **`--code -` reads from stdin deliberately.** Passing the code as a flag value works but warns:\n  a value on the command line is visible to other processes through `ps` and is saved in shell\n  history.\n- **The two legs must run on the same machine, under the same configuration directory.** Leg one\n  writes a short-lived resume file beside the config file, readable only by you, holding the session\n  id and the private key that opens the delivered credential. Leg two reads it and deletes it. Split\n  the legs across machines, users, containers or config-directory overrides and the exchange cannot\n  complete.\n- **`--code -` with no setup in progress exits `2`** and says so. Run leg one first.\n- The code is short, case-insensitive, and drops anything that is not one of its own characters, so\n  a pasted code with stray spaces or punctuation is still accepted.\n\n### Human shape\n\nWith stdout on a terminal and no `--json`, `setup` prints the link, opens a browser, and prompts for\nthe code (up to three attempts). `--no-browser` prints the link rather than opening it, which is what\nyou want over SSH or anywhere a browser would open on the wrong machine.\n\nOne asymmetry to know before scripting it: **stdout on a terminal with stdin piped is refused at exit\n`2`**, because there is nowhere to paste a code into. It names `CURVIATE_API_KEY` and\n`curviate login --api-key -` as the alternatives. A *fully* piped invocation is not that case: it is\nthe agent shape above, and it exits `0`.\n\n### What each failure means\n\n| Exit | Cause | What to do |\n|---|---|---|\n| `0` | Authenticated. The key is saved to the named profile. | Continue to `doctor`. |\n| `2` | Usage: no setup in progress, or a terminal was needed and there was none. | Fix the invocation. Never retry unchanged. |\n| `3` | The code was wrong, expired, or the attempts ran out; or the workspace has no API key to deliver; or the key saved but a verifying call did not succeed. **Read the message: these are deliberately four different sentences, and only the first is fixed by starting over.** | Wrong or expired code: run `setup` again for a fresh one. No API key: create one in the dashboard first. Saved but unverified: run `doctor`. |\n| `6` | Too many attempts against this endpoint. | Wait, then try again. |\n| `7` | The API could not be reached. | Check the network, and the base URL if you overrode it. |\n| `1` | The exchange failed in a way the CLI did not expect, or the delivered credential could not be opened by this process. | Worth one retry. If it repeats it is a bug to report, not a state to work around. |\n\nA code is short-lived and allows only a handful of attempts. If in doubt, start a fresh `setup`\nrather than reusing anything.\n\n## 3. Confirm: `curviate doctor`\n\n```bash\ncurviate doctor --json\n```\n\nIt reports the CLI version, the config path, the active profile, the base URL, where the credential\nresolved from, **which workspace it authenticated as**, whether the API is reachable, whether the\ncredential is accepted, and every connected account with its status.\n\nThree things that surprise people:\n\n- **Run `setup` first, then `doctor`, not the other way round.** Reachability is only tested through\n  a real call, so with no credential resolved `doctor` reports the API as unreachable. That is not a\n  network problem; there was nothing to call with.\n- **The workspace line is populated only for a credential `setup` wrote.** A key supplied through the\n  environment or `login` authenticates perfectly well, but `doctor` has no name to show for it and\n  says so. And where a name is shown it identifies the workspace; it is not promised to be a\n  friendly, human-chosen label.\n- **Zero connected accounts is a pass, not a failure.** `doctor` reports `0 connected` and exits `0`.\n  That is the expected state of a workspace nobody has connected an account to yet.\n\n`doctor` reports rather than throws, so it prints a report and not an error envelope. Branch on the\nexit code and read `checks[]` for which stage failed:\n\n- **`0`**: every check passed.\n- **`2`**: the request was refused before it was sent, so nothing reached the network. An empty\n  credential, or a malformed base URL (the report's detail reads `Invalid base URL: expected an\n  absolute http:\u002F\u002F or https:\u002F\u002F URL`). Fix the invocation or the value; a retry cannot help.\n- **`3`**: no credential resolved, or one resolved and was rejected. Run `setup`.\n- **`7`**: the API could not be reached, or was reached and answered with a platform fault. Both are\n  worth a retry.\n\n**This list is not exhaustive, and the codes beyond it are not `doctor`'s own.** The credential check\nis a real API call, and whatever refusal comes back is passed straight through, so any code the API\ncan produce can surface here. The one a first run meets most often is **exit `5` on a workspace with\nno active seat**: the credential is fine and the network is fine, and no amount of re-running `setup`\nwill change it. Look an unfamiliar code up in the area skill for the surface you are about to use:\neach ends with a table of the codes its commands produce and the action each implies.\n\nThe `checks[]` entries separate the three stages, and the wording is exact: `api reachable` reading\n`not checked` means the request never left this machine, while `could not reach` means it left and\nfound nothing. `credential valid` reading `not checked` means nothing asked the credential anything;\nit is not a verdict on the credential, so do not go looking for a bad key on the strength of it.\n\n## 4. Connect a LinkedIn account\n\nCurviate acts on LinkedIn through an account you connect. Until one is connected, authentication is\ncomplete and correct while **every account-scoped command is unusable**: that is the whole state,\nand it is not an error to be debugged.\n\n```bash\ncurviate account list --json    # empty on a fresh workspace\n```\n\nIf `setup` reported an `account_id`, an account is already connected **and already set as this\nprofile's default**; skip the rest of this step, including the block below, and go to step 5.\n\nOtherwise, find a free seat first:\n\n```bash\ncurviate account seats --json    # seat_id of any entry with \"occupied\": false\n```\n\nThen connect one with `curviate account link --seat-id \u003Cid> ...`. It attaches a LinkedIn account to\nthat seat and usually needs a verification code, which means a human. Non-interactive shells get exit\n`12` and finish through the checkpoint flow. **The command surface, its flags and the checkpoint\ncommands are in `curviate-profile`**: read that before running it, because a half-finished connect\nleaves a session to poll rather than a clean failure.\n\nThen tell the CLI which account to act as:\n\n```bash\ncurviate account list --json              # the acc_id of the account you just connected\ncurviate config set-account \u003Cacc_id>      # make it this profile's default\n```\n\n**Do not skip this.** `setup` stores a default account only when the workspace already had one to\nhand back, which is exactly what the expected first run does not have. Without a default, every\naccount-scoped command, including the demonstration below, stops at exit `2` asking for `--account`.\nSet it once here, or pass `--account \u003Cacc_id>` on each call. **`config set-account` writes to the\nstored profile, so it needs one**: if your credential came from `CURVIATE_API_KEY` rather than\n`setup`, it exits `2` saying there is no config file, and `--account \u003Cacc_id>` per call (or\n`CURVIATE_ACCOUNT` in the environment) is the path for you. Account ids change when an account is\nreconnected, so read one live rather than hard-coding it.\n\n## 5. Prove the whole path: the same read, twice\n\nThis is the demonstration, and it is worth running because it exercises authentication, the\nconnected account, projection and the store in one pass. Both calls are reads of your own profile.\n\n```bash\ncurviate profile me --mode live --json          # \"source\":\"live\"\ncurviate profile me --mode cache_only --json    # \"source\":\"store\", with \"observed_at\"\n```\n\nThe first forces a live call to LinkedIn. The second refuses to go to LinkedIn at all and answers\nfrom the copy the first one stored, so it comes back with `source: store` and an `observed_at`\ntimestamp saying when that copy was taken.\n\n**Read `source` rather than assuming it.** That field, not the freshness you asked for, is what\nactually happened. The full set of modes, and what a `cache_only` read does when nothing is stored,\nare in `curviate-profile`.\n\n## Full command surface\n\n\u003C!-- generated: command surface, CLI 0.33.0 -->\n\nRead from the CLI's own `--help` at version 0.33.0. Descriptions, traps and confidence\ntags elsewhere in this skill are hand-written and carry the version they were established against.\n\nEvery command below that takes flags at all also accepts `--base-url`, `--json`, `--profile`.\n\n| Command | Arguments | Flags |\n|---|---|---|\n| `curviate setup` | *(none)* | `--no-browser`, `--code` |\n| `curviate doctor` | *(none)* | `--api-key`, `--timeout` |\n| `curviate login` | *(none)* | `--api-key`, `--account` |\n\n\u003C!-- \u002Fgenerated -->\n\n## Where to go next\n\nBack to `curviate` for the dispatch table. The common first tasks: look someone up\n(`curviate-profile`), find a set of people (`curviate-search`), read your inbox (`curviate-inbox`).\nEach area skill ends with the exit codes its commands can produce and what each one implies; branch\non those, never on the message text.\n",{"name":5,"license":15,"metadata":55,"description":7},{"tags":56,"author":62,"version":63,"repository":64},[57,58,59,60,40,61],"LinkedIn","CLI","Agents","Sales","Outreach","Curviate","0.1.0","https:\u002F\u002Fgithub.com\u002FCurviate\u002Fcurviate-plugin","main","2b749c82b5d11fac39d9a6e021ce3f5b8a541bd1","cli-tool\u002Fcomponents\u002Fskills\u002Fcurviate\u002Fcurviate-quickstart",3911,"https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates","https:\u002F\u002Faitmpl.com","https:\u002F\u002Fgithub.com\u002Fdavila7","https:\u002F\u002Fdanielavila.me","Building AI dev tools with LLMs","2026-11-11 21:38:31",{"source":8,"skillName":5,"commands":76,"prompt":86,"zipAvailable":87},[77,80,83],{"tool":78,"command":79},"npx","npx skills add davila7\u002Fclaude-code-templates --skill curviate-quickstart",{"tool":81,"command":82},"bunx","bunx skills add davila7\u002Fclaude-code-templates --skill curviate-quickstart",{"tool":84,"command":85},"pnpm","pnpm dlx skills add davila7\u002Fclaude-code-templates --skill curviate-quickstart","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fcurviate\u002Fcurviate-quickstart\u002FSKILL.md",true,[],[90,116,143,175,210,234],{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":97,"publishedAt":50},"01M70NHBTQ8BV0A2SCAZXEPHCX","wealthville-pool-scoring","Score DeFi liquidity pools before providing liquidity. Returns a 0-100 pool score and an Enter\u002FHold\u002FExit\u002FReduce\u002FAvoid verdict with calibrated confidence for ~68,800 Solana pools and 575 EVM pools, plus a public miss-inclusive track record. Use when asked whether a pool is worth LPing into, which pools look best right now, or how accurate the scores have actually been.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fweb-data\u002Fwealthville-pool-scoring\u002FSKILL.md",[96],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[98,99,100,101,102,103,104,105,106,27,107,108,109,110,111,112,113,39,114,115],"scoring","pool","wealthville pool scoring","liquidity","which","for","exit","now","use","reduce","with","record","track","verdict","evm","solana","public","look",{"skillNo":117,"name":118,"nameZh":6,"description":119,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":120,"categories":121,"keywords":123,"publishedAt":50},"01M70MB6VR8ERJ4EM8D5KHBMEV","technical-debt-auditor","Scans a codebase and produces a quantified technical debt report with a prioritized remediation plan. Analyzes dead code, TODO\u002FFIXME\u002FHACK markers, dependency staleness, complexity hotspots, code duplication signals, test coverage gaps, unused exports, and outdated patterns. Use this skill when the user asks to audit their codebase for tech debt, wants a health check of their code quality, says things like \"where should I focus refactoring\", \"find code smells\", \"audit my codebase\", \"what technical debt do we have\", \"prioritize cleanup work\", or asks about code quality metrics. Also trigger when the user mentions \"code rot\", \"legacy cleanup\", \"refactoring priorities\", or wants to understand the overall health of a project before a major effort. Trigger even if the user doesn't say \"technical debt\" explicitly — any request to systematically assess code quality and identify improvement priorities belongs here.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdevelopment\u002Ftechnical-debt-auditor\u002FSKILL.md",[122],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[124,125,126,127,128,129,130,131,132,133,134,27,135,136,137,138,139,140,141,142],"technical debt","debt","technical","technical debt auditor","auditor","coverage","complexity","markers","project","cleanup","unused","code","todo","work","plan","focus","signals","audit","dependency",{"skillNo":144,"name":145,"nameZh":6,"description":146,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":147,"categories":148,"keywords":158,"publishedAt":50},"01M70KPYH9QNHCNME5GJQGE22K","readonly-sql-guard","Make an agent's database access provably read-only. Use when an agent runs caller-supplied SQL against a production\u002FERP\u002Fanalytics database (a \"query\" MCP tool, text-to-SQL, an admin console), or when reviewing DB-access code for write\u002FDDL\u002Fside-effect\u002Ftransaction-escape risks. Checks what a statement CALLS, not just its shape.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fdatabase\u002Freadonly-sql-guard\u002FSKILL.md",[149,152,153],{"categoryNo":150,"categoryName":151,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":154,"categoryName":155,"taxonomyType":156,"taxonomyTypeDesc":157},"01JBSEEDCAT000000000000033","数据分析师","PROFESSION","适用职业",[159,160,161,162,163,164,165,103,166,167,168,169,170,35,171,106,172,173,27,174],"readonly","guard","sql guard","sql","readonly sql guard","erp","effect","read-only","db","just","tool","calls","MCP","access","database","query",{"skillNo":176,"name":177,"nameZh":6,"description":178,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":179,"licenseFileUrl":180,"sourceHtmlUrl":181,"categories":182,"keywords":186,"publishedAt":50},"01M6AFY6F7CC2BBZP1QD6VTNVC","web-artifacts-builder","Suite of tools for creating elaborate, multi-component claude.ai HTML artifacts using modern frontend web technologies (React, Tailwind CSS, shadcn\u002Fui). Use for complex artifacts requiring state management, routing, or shadcn\u002Fui components - not for simple single-file HTML\u002FJSX artifacts.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Futilities\u002Fweb-artifacts-builder\u002FSKILL.md",[183],{"categoryNo":184,"categoryName":185,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000009","前端开发",[187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,103,209],"artifacts builder","web","web artifacts builder","artifacts","builder","state","routing","state management","component","css","frontend","ui","Tools","file","components","shadcn","react","Tailwind","suite","jsx","html","single-file","ui components",{"skillNo":211,"name":212,"nameZh":6,"description":213,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":179,"licenseFileUrl":214,"sourceHtmlUrl":215,"categories":216,"keywords":221,"publishedAt":50},"01M6AFWF33SFV9Z3SF4PEEE4BF","webapp-testing","Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fsecurity\u002Fwebapp-testing\u002FSKILL.md",[217,220],{"categoryNo":218,"categoryName":219,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000008","测试与质量",{"categoryNo":184,"categoryName":185,"taxonomyType":21,"taxonomyTypeDesc":22},[222,223,224,225,226,188,227,197,198,228,229,230,231,103,232,108,233],"webapp testing","testing","webapp","browser","playwright","applications","debugging","toolkit","logs","screenshots","local","behavior",{"skillNo":235,"name":236,"nameZh":6,"description":237,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":179,"licenseFileUrl":238,"sourceHtmlUrl":239,"categories":240,"keywords":248,"publishedAt":50},"01M6AF67Q67JJJ5V51JR2P52KA","discernment-nudge","After you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fdavila7\u002Fclaude-code-templates\u002Fblob\u002Fmain\u002Fcli-tool\u002Fcomponents\u002Fskills\u002Fproductivity\u002Fdiscernment-nudge\u002FSKILL.md",[241,244,245],{"categoryNo":242,"categoryName":243,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},{"categoryNo":246,"categoryName":247,"taxonomyType":156,"taxonomyTypeDesc":157},"01JBSEEDCAT000000000000036","内容创作者",[249,250,251,252,45,253,254,255,190,256,200,135,257,258,259,260,34,35,261,27,262,263,264,265,266,39,267],"discernment nudge","nudge","interpretation","help","reasoning","data","writing","lookup","content","review","conversation","short","invoke","you","analysis","claims","argument","assemble","recommendations"]