[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6MMD41C3SWCC8SS0983QHS4":3,"skill-same:01M6MMD41C3SWCC8SS0983QHS4":77},{"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":41,"linkOnly":42,"markdownContent":43,"markdownBody":44,"frontMatter":45,"refName":48,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":53,"ownerHtmlUrl":54,"ownerBlogUrl":55,"ownerDescription":56,"updatedAt":57,"install":58,"resources":70},"01M6MMD41C3SWCC8SS0983QHS4","runtype-sdk-marathon",null,"Build source-controlled Runtype workflows with the SDK or CLI: config-as-code deploys from CI (ensure), save-and-run flows (upsert), long-running or multi-session agent runs with Marathon (runtype marathon or runtype agents task), and Marathon playbooks.","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1554,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fruntypelabs\u002Fskills\u002Fskills\u002Fruntype-sdk-marathon\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40],"sdk","runtype","runtype sdk marathon","running","cli","source","true","with","flows","multi","long","as","multi-session","task","build","session","save","ci","code","run","config","workflows","2026-11-10 10:12:30",false,"---\nname: runtype-sdk-marathon\ndescription: 'Build source-controlled Runtype workflows with the SDK or CLI: config-as-code deploys from CI (ensure), save-and-run flows (upsert), long-running or multi-session agent runs with Marathon (runtype marathon or runtype agents task), and Marathon playbooks.'\nuser-invocable: true\nargument-hint: '[SDK, CLI, or Marathon task]'\n---\n\n# Runtype SDK and Marathon\n\nUse this skill for code-first Runtype workflows, the CLI, and Marathon. Use\n`runtype-build-product` when the user designs hosted product resources through MCP.\n\n## SDK guidance\n\nFetch live docs when available:\n\n- `get_platform_documentation(topic=\"sdk-reference\")`\n- `get_platform_documentation(topic=\"types-flow-steps\")`\n- `get_platform_documentation(topic=\"types-entities\")`\n- `get_platform_documentation(topic=\"flow-step-types\")`\n\nCode-first modes:\n\n- **Stored**: create persistent agents or flows in Runtype.\n- **Upsert**: running the code creates the flow if it does not exist and updates it when\n  its steps change. It keeps a version of each change by default\n  (`createVersionOnChange: false` turns that off). If the flow was last edited in the\n  dashboard or through the API, upsert fails unless you set\n  `allowOverwriteExternalChanges: true`.\n- **Ensure (config-as-code)**: converge a definition from your repo onto Runtype at deploy\n  time without running it. Ensure never deletes a resource, and agents, flows, and skills\n  keep a version of each change. The SDK has `ensure` for flows, agents, tools, skills,\n  evals, products, surfaces, and client tokens; `evals.ensure` replaces the suite's cases.\n  The CLI has `runtype agents ensure \u003Cfile>` for a JSON agent definition; add\n  `--expect-no-changes` to fail a CI job on drift.\n- **Virtual**: the definition is supplied per execution without creating a saved agent or\n  flow. Execution logs, traces, and retention policies still apply, so this is not a\n  zero-retention mode.\n\nUse ensure for CI\u002FCD deploys and upsert for save-and-run scripts. Prefer stored, upsert, or\nensure for production workflows, because dashboard inspection, logs, evals, and versioning\nare easier. Use virtual for tests, one-offs, or temporary generated flows. For\nprivacy-sensitive workloads, verify retention separately.\n\nUse local tools when a tool must run in the user's browser or server: pass a map of tool\nnames to async functions to `runWithLocalTools`. Use hidden parameters when auth context,\ntenant IDs, or sensitive request data must not appear in the tool schema the model sees:\nlist them in the tool's `hiddenParameterNames`, and the runtime fills them from execution\nvariables. Names that start with `_internal` are never filled; use `{{secret:NAME}}`\nreferences for credentials. The `sdk-reference` topic has the signatures.\n\n## CLI setup\n\nInstall or run directly:\n\n```bash\nnpm install -g @runtypelabs\u002Fcli\nnpx @runtypelabs\u002Fcli@latest \u003Ccommand>\n```\n\nCheck auth before you change it:\n\n```bash\nruntype auth status\n```\n\n`runtype auth status` checks only the stored login or signup state. Then pick the first\ncase that applies:\n\n1. `RUNTYPE_API_KEY` is set: run `runtype auth whoami --no-tty` and reuse that account.\n   This is also how CLI commands authenticate in CI.\n2. A stored account is authenticated: reuse it.\n3. A signup is pending: run the `next` command that `auth status` prints.\n4. `auth status` reports `unauthenticated` with a `requestedOrgId`: the project is linked\n   (`.runtype\u002Fproject.json`) to an organization with no stored login. Do not register; have\n   the user privately run `runtype auth login --api-key \u003Ckey>` for that organization.\n5. The user needs a new account: run `runtype auth register --email \u003Cemail>`, then\n   `runtype auth verify \u003Ccode>`.\n6. The user wants the browser login flow of `runtype auth login`: the user runs it in their\n   own interactive terminal, not in an automated agent's shell.\n\nFor existing-account or installation recovery, follow\n`https:\u002F\u002Fruntype.ai\u002F.well-known\u002Fagent.md`. Never ask for API keys in chat. Have the user\nconfigure credentials privately.\n\nCommon commands include `runtype agents list`, `runtype agents ensure`, `runtype dispatch`,\n`runtype flows create`, `runtype records create`, `runtype schedules`, `runtype models`,\n`runtype batch`, `runtype eval`, `runtype persona`, `runtype products init`, and\n`runtype validate-product`. For flags not listed here, run `runtype --help` or\n`runtype \u003Ccommand> --help`.\n\n## Marathon\n\nMarathon is a CLI command that runs long research, code-editing, and build tasks across\nmultiple sessions. It streams progress, carries context from one session to the next,\nsupports playbooks, and can use sandboxes. `runtype marathon` is an alias for\n`runtype agents task`. The `get_platform_documentation` topics do not cover Marathon, so use\n`runtype marathon --help` as the flag reference. User docs:\n`https:\u002F\u002Fdocs.runtype.com\u002Fuser-guide\u002Fagents\u002Fmarathon-long-running-tasks`.\n\nThe first argument is an agent ID or name. If no agent matches, the CLI creates an agent\nwith that name in the user's account.\n\n```bash\nruntype marathon researcher \\\n  --goal \"Research recent AI announcements and summarize them\" \\\n  --tools firecrawl \\\n  --max-sessions 5 \\\n  --max-cost 2\n\nruntype marathon \"Code Editor\" \\\n  --goal \"Refactor the auth module and run tests\" \\\n  --max-cost 10\n\nruntype marathon calculator \\\n  --goal \"Build a calculator in 3D and deploy it publicly\" \\\n  --sandbox daytona\n```\n\nGuardrails:\n\n- **Cost and length**: `--max-sessions` defaults to 50 and there is no default budget. For\n  unattended runs, always set `--max-cost \u003Cusd>` and an explicit `--max-sessions \u003Cn>`.\n- **Local files**: local file tools such as `read_file`, `write_file`, and\n  `list_directory` are on and can read and write anywhere under the current working\n  directory. The `--no-local-tools` flag does not currently turn them off, so start\n  research-only runs from an empty directory, and never from an untrusted one.\n- **Unattended runs**: without a TTY, or with `--json`, the interactive startup screen is\n  off. `--json` prints the final result as JSON. Checkpoint pauses auto-continue after\n  `--checkpoint-timeout` seconds (default 10). The `--no-checkpoint` flag does not\n  currently remove the pauses.\n\nBuilt-in tools you can enable with `--tools` include `exa`, `firecrawl`, `dalle`,\n`openai_web_search`, `anthropic_web_search`, and `google_search`. The CLI validates tool IDs\nand model compatibility at startup.\n\n`--sandbox` accepts `daytona`, `runtype-sandbox`, `cloudflare-worker`, or `quickjs`. Only\n`daytona` and `runtype-sandbox` can deploy to a public preview URL.\n\nModels: `--model` sets the model for the run, `--planning-model` and `--execution-model`\nset it per phase, and `--fallback-model` is used when the primary model fails. Use routed\nmodel IDs such as `claude-sonnet-5` or `claude-haiku-4-5`. For `--max-tokens` and\n`--temperature`, a playbook milestone value beats the CLI flag, and the CLI flag beats the\nplaybook-level default.\n\nTo watch a run in the dashboard, add `--track`. Marathon syncs progress to a Runtype record\nnamed after the task.\n\n### Resume or restart a run\n\nMarathon saves local state per task, by default under\n`~\u002F.runtype\u002Fprojects\u002F\u003Chash>\u002Fmarathons\u002F` (change it with `--state-dir`).\n\n- `--resume \"\u003Cnew instruction>\"` continues from saved state, for example after the run hits\n  its session or cost cap. The message is optional.\n- `--session \u003Cname>` resumes a specific session.\n- `--fresh` starts a new run and ignores saved state.\n- `--name \u003Cname>` keeps separate tasks for the same agent apart. It defaults to the agent\n  name.\n\nFor long runs that hit context limits, adjust `--compact-threshold`,\n`--tool-context observation-mask`, or `--offload-threshold`. See `runtype marathon --help`.\n\n## Playbooks\n\nUse a playbook for repeatable multi-phase work such as research, build, verify, and polish.\nLoad one with `--playbook`:\n\n```bash\nruntype marathon researcher \\\n  --goal \"Research the topic, then build a demo\" \\\n  --playbook research-build \\\n  --max-cost 5\n```\n\n`--playbook NAME` accepts a file path, or a name that resolves to\n`.runtype\u002Fmarathons\u002Fplaybooks\u002FNAME.{yaml,yml,json,ts,mts}`. The CLI checks the current\ndirectory first, then the home directory. The CLI ships no built-in playbooks, so\n`research-build` above is a file the user writes.\n\nA playbook has a `name` and a list of `milestones`. Each milestone has a `name`,\n`instructions`, and optional `model`, `fallbackModels`, `completionCriteria`, `maxTokens`,\nand `temperature`. Top-level fields include `description`, `rules`, `verification`,\n`policy`, `fallbackOnEmpty`, `maxTokens`, `temperature`, and `plugins`. Keep `rules`\nexplicit about file scope, verification, and deployment expectations.\n\nA TypeScript playbook default-exports a config, or a factory that receives\n`{ registerWorkflowHook }`. Wrap it in `definePlaybook` from `@runtypelabs\u002Fsdk` for types:\n\n```ts\nimport { definePlaybook } from '@runtypelabs\u002Fsdk'\n\nexport default definePlaybook({\n  name: 'build-only',\n  milestones: [\n    {\n      name: 'build',\n      instructions: 'Build the feature described in the goal, then run the tests.',\n      canAcceptCompletion: true,\n    },\n  ],\n})\n```\n\nTypeScript playbooks and `plugins` modules run as code with the user's privileges when the\nplaybook loads. Load only playbooks and plugins the user trusts.\n\nIf the `runtype` skill is also installed, its references cover working modes in more depth.\nPrefer live MCP documentation over local reference files.\n","# Runtype SDK and Marathon\n\nUse this skill for code-first Runtype workflows, the CLI, and Marathon. Use\n`runtype-build-product` when the user designs hosted product resources through MCP.\n\n## SDK guidance\n\nFetch live docs when available:\n\n- `get_platform_documentation(topic=\"sdk-reference\")`\n- `get_platform_documentation(topic=\"types-flow-steps\")`\n- `get_platform_documentation(topic=\"types-entities\")`\n- `get_platform_documentation(topic=\"flow-step-types\")`\n\nCode-first modes:\n\n- **Stored**: create persistent agents or flows in Runtype.\n- **Upsert**: running the code creates the flow if it does not exist and updates it when\n  its steps change. It keeps a version of each change by default\n  (`createVersionOnChange: false` turns that off). If the flow was last edited in the\n  dashboard or through the API, upsert fails unless you set\n  `allowOverwriteExternalChanges: true`.\n- **Ensure (config-as-code)**: converge a definition from your repo onto Runtype at deploy\n  time without running it. Ensure never deletes a resource, and agents, flows, and skills\n  keep a version of each change. The SDK has `ensure` for flows, agents, tools, skills,\n  evals, products, surfaces, and client tokens; `evals.ensure` replaces the suite's cases.\n  The CLI has `runtype agents ensure \u003Cfile>` for a JSON agent definition; add\n  `--expect-no-changes` to fail a CI job on drift.\n- **Virtual**: the definition is supplied per execution without creating a saved agent or\n  flow. Execution logs, traces, and retention policies still apply, so this is not a\n  zero-retention mode.\n\nUse ensure for CI\u002FCD deploys and upsert for save-and-run scripts. Prefer stored, upsert, or\nensure for production workflows, because dashboard inspection, logs, evals, and versioning\nare easier. Use virtual for tests, one-offs, or temporary generated flows. For\nprivacy-sensitive workloads, verify retention separately.\n\nUse local tools when a tool must run in the user's browser or server: pass a map of tool\nnames to async functions to `runWithLocalTools`. Use hidden parameters when auth context,\ntenant IDs, or sensitive request data must not appear in the tool schema the model sees:\nlist them in the tool's `hiddenParameterNames`, and the runtime fills them from execution\nvariables. Names that start with `_internal` are never filled; use `{{secret:NAME}}`\nreferences for credentials. The `sdk-reference` topic has the signatures.\n\n## CLI setup\n\nInstall or run directly:\n\n```bash\nnpm install -g @runtypelabs\u002Fcli\nnpx @runtypelabs\u002Fcli@latest \u003Ccommand>\n```\n\nCheck auth before you change it:\n\n```bash\nruntype auth status\n```\n\n`runtype auth status` checks only the stored login or signup state. Then pick the first\ncase that applies:\n\n1. `RUNTYPE_API_KEY` is set: run `runtype auth whoami --no-tty` and reuse that account.\n   This is also how CLI commands authenticate in CI.\n2. A stored account is authenticated: reuse it.\n3. A signup is pending: run the `next` command that `auth status` prints.\n4. `auth status` reports `unauthenticated` with a `requestedOrgId`: the project is linked\n   (`.runtype\u002Fproject.json`) to an organization with no stored login. Do not register; have\n   the user privately run `runtype auth login --api-key \u003Ckey>` for that organization.\n5. The user needs a new account: run `runtype auth register --email \u003Cemail>`, then\n   `runtype auth verify \u003Ccode>`.\n6. The user wants the browser login flow of `runtype auth login`: the user runs it in their\n   own interactive terminal, not in an automated agent's shell.\n\nFor existing-account or installation recovery, follow\n`https:\u002F\u002Fruntype.ai\u002F.well-known\u002Fagent.md`. Never ask for API keys in chat. Have the user\nconfigure credentials privately.\n\nCommon commands include `runtype agents list`, `runtype agents ensure`, `runtype dispatch`,\n`runtype flows create`, `runtype records create`, `runtype schedules`, `runtype models`,\n`runtype batch`, `runtype eval`, `runtype persona`, `runtype products init`, and\n`runtype validate-product`. For flags not listed here, run `runtype --help` or\n`runtype \u003Ccommand> --help`.\n\n## Marathon\n\nMarathon is a CLI command that runs long research, code-editing, and build tasks across\nmultiple sessions. It streams progress, carries context from one session to the next,\nsupports playbooks, and can use sandboxes. `runtype marathon` is an alias for\n`runtype agents task`. The `get_platform_documentation` topics do not cover Marathon, so use\n`runtype marathon --help` as the flag reference. User docs:\n`https:\u002F\u002Fdocs.runtype.com\u002Fuser-guide\u002Fagents\u002Fmarathon-long-running-tasks`.\n\nThe first argument is an agent ID or name. If no agent matches, the CLI creates an agent\nwith that name in the user's account.\n\n```bash\nruntype marathon researcher \\\n  --goal \"Research recent AI announcements and summarize them\" \\\n  --tools firecrawl \\\n  --max-sessions 5 \\\n  --max-cost 2\n\nruntype marathon \"Code Editor\" \\\n  --goal \"Refactor the auth module and run tests\" \\\n  --max-cost 10\n\nruntype marathon calculator \\\n  --goal \"Build a calculator in 3D and deploy it publicly\" \\\n  --sandbox daytona\n```\n\nGuardrails:\n\n- **Cost and length**: `--max-sessions` defaults to 50 and there is no default budget. For\n  unattended runs, always set `--max-cost \u003Cusd>` and an explicit `--max-sessions \u003Cn>`.\n- **Local files**: local file tools such as `read_file`, `write_file`, and\n  `list_directory` are on and can read and write anywhere under the current working\n  directory. The `--no-local-tools` flag does not currently turn them off, so start\n  research-only runs from an empty directory, and never from an untrusted one.\n- **Unattended runs**: without a TTY, or with `--json`, the interactive startup screen is\n  off. `--json` prints the final result as JSON. Checkpoint pauses auto-continue after\n  `--checkpoint-timeout` seconds (default 10). The `--no-checkpoint` flag does not\n  currently remove the pauses.\n\nBuilt-in tools you can enable with `--tools` include `exa`, `firecrawl`, `dalle`,\n`openai_web_search`, `anthropic_web_search`, and `google_search`. The CLI validates tool IDs\nand model compatibility at startup.\n\n`--sandbox` accepts `daytona`, `runtype-sandbox`, `cloudflare-worker`, or `quickjs`. Only\n`daytona` and `runtype-sandbox` can deploy to a public preview URL.\n\nModels: `--model` sets the model for the run, `--planning-model` and `--execution-model`\nset it per phase, and `--fallback-model` is used when the primary model fails. Use routed\nmodel IDs such as `claude-sonnet-5` or `claude-haiku-4-5`. For `--max-tokens` and\n`--temperature`, a playbook milestone value beats the CLI flag, and the CLI flag beats the\nplaybook-level default.\n\nTo watch a run in the dashboard, add `--track`. Marathon syncs progress to a Runtype record\nnamed after the task.\n\n### Resume or restart a run\n\nMarathon saves local state per task, by default under\n`~\u002F.runtype\u002Fprojects\u002F\u003Chash>\u002Fmarathons\u002F` (change it with `--state-dir`).\n\n- `--resume \"\u003Cnew instruction>\"` continues from saved state, for example after the run hits\n  its session or cost cap. The message is optional.\n- `--session \u003Cname>` resumes a specific session.\n- `--fresh` starts a new run and ignores saved state.\n- `--name \u003Cname>` keeps separate tasks for the same agent apart. It defaults to the agent\n  name.\n\nFor long runs that hit context limits, adjust `--compact-threshold`,\n`--tool-context observation-mask`, or `--offload-threshold`. See `runtype marathon --help`.\n\n## Playbooks\n\nUse a playbook for repeatable multi-phase work such as research, build, verify, and polish.\nLoad one with `--playbook`:\n\n```bash\nruntype marathon researcher \\\n  --goal \"Research the topic, then build a demo\" \\\n  --playbook research-build \\\n  --max-cost 5\n```\n\n`--playbook NAME` accepts a file path, or a name that resolves to\n`.runtype\u002Fmarathons\u002Fplaybooks\u002FNAME.{yaml,yml,json,ts,mts}`. The CLI checks the current\ndirectory first, then the home directory. The CLI ships no built-in playbooks, so\n`research-build` above is a file the user writes.\n\nA playbook has a `name` and a list of `milestones`. Each milestone has a `name`,\n`instructions`, and optional `model`, `fallbackModels`, `completionCriteria`, `maxTokens`,\nand `temperature`. Top-level fields include `description`, `rules`, `verification`,\n`policy`, `fallbackOnEmpty`, `maxTokens`, `temperature`, and `plugins`. Keep `rules`\nexplicit about file scope, verification, and deployment expectations.\n\nA TypeScript playbook default-exports a config, or a factory that receives\n`{ registerWorkflowHook }`. Wrap it in `definePlaybook` from `@runtypelabs\u002Fsdk` for types:\n\n```ts\nimport { definePlaybook } from '@runtypelabs\u002Fsdk'\n\nexport default definePlaybook({\n  name: 'build-only',\n  milestones: [\n    {\n      name: 'build',\n      instructions: 'Build the feature described in the goal, then run the tests.',\n      canAcceptCompletion: true,\n    },\n  ],\n})\n```\n\nTypeScript playbooks and `plugins` modules run as code with the user's privileges when the\nplaybook loads. Load only playbooks and plugins the user trusts.\n\nIf the `runtype` skill is also installed, its references cover working modes in more depth.\nPrefer live MCP documentation over local reference files.\n",{"name":5,"description":7,"argument-hint":46,"user-invocable":47},"[SDK, CLI, or Marathon task]",true,"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002Fruntypelabs\u002Fskills\u002Fskills\u002Fruntype-sdk-marathon",468,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-11 02:22:54",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":47},[60,63,66],{"tool":61,"command":62},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill runtype-sdk-marathon",{"tool":64,"command":65},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill runtype-sdk-marathon",{"tool":67,"command":68},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill runtype-sdk-marathon","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fruntypelabs\u002Fskills\u002Fskills\u002Fruntype-sdk-marathon\u002FSKILL.md",[71],{"path":72,"resourceType":73,"resourceTypeDesc":74,"sizeBytes":75,"sourceHtmlUrl":76},"agents\u002Fopenai.yaml","REFERENCE","参考文档",527,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fruntypelabs\u002Fskills\u002Fskills\u002Fruntype-sdk-marathon\u002Fagents\u002Fopenai.yaml",[78,104,127,161,182,196],{"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":41},"01M716A6ETNB9B1SQ3GXZC75VR","deja-search","deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-search\u002FSKILL.md",[],[85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,23,102,103],"search","deja search","error","ready","never","memory","re","decide","fix","design","debugging","when","you","done","reviewer","io","change","merge","past",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":110,"publishedAt":41},"01M716A3J2W5SENSE9VGMXB375","deja-reflect","End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-reflect\u002FSKILL.md",[],[111,112,113,114,90,115,116,34,117,118,119,120,121,100,26,122,123,124,125,126],"reflect","deja reflect","brew","remember","project rules","project","end","reflection","Rules","decisions","review","record","github","next","https","node",{"skillNo":128,"name":129,"nameZh":6,"description":130,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":131,"categories":132,"keywords":144,"publishedAt":41},"01M715T3RR4BAYSR57HPPTJ0PF","revid","Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftmakersas\u002Frevid-skills\u002Fskills\u002Frevid\u002FSKILL.md",[133,138,141],{"categoryNo":134,"categoryName":135,"taxonomyType":136,"taxonomyTypeDesc":137},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":139,"categoryName":140,"taxonomyType":136,"taxonomyTypeDesc":137},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":142,"categoryName":143,"taxonomyType":136,"taxonomyTypeDesc":137},"01JBSEEDCAT000000000000023","Skill与智能体开发",[129,145,146,147,148,149,116,150,91,151,152,153,154,96,155,156,157,158,159,160],"Video","short-form","standalone","PDF","viral","tiktok","short","social","MCP","song","URL","image","mcp server","schedule","api","publish",{"skillNo":162,"name":163,"nameZh":6,"description":164,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":165,"categories":166,"keywords":167,"publishedAt":41},"01M71569ZM2X3H5RBFK8PVKMK2","review-study","Review a QuestionPunk survey before launch for wording, answer choices, branching, and missing settings. Use for check my survey or review this draft before publishing.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Freview-study\u002FSKILL.md",[],[168,169,121,170,171,172,173,174,175,176,177,178,179,180,181],"study","review study","for","answer","missing","survey","publishing","use","check","draft","settings","launch","my","branching",{"skillNo":183,"name":184,"nameZh":6,"description":185,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":186,"categories":187,"keywords":188,"publishedAt":41},"01M71567GV0C35YQBY0DPGR86B","manage-study","Find QuestionPunk studies, check status, publish an explicitly requested draft, or retrieve its participant link. Use for find my survey, publish this study, or how many people responded.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fmanage-study\u002FSKILL.md",[],[189,168,190,170,173,175,191,176,177,160,192,193,194,195,180],"manage study","manage","people","how","find","link","status",{"skillNo":197,"name":198,"nameZh":6,"description":199,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":200,"categories":201,"keywords":202,"publishedAt":41},"01M71565A8TCA01AZQZRGRYMFZ","create-study","Create a saved QuestionPunk survey draft from a research brief or edit specified questions. Use for requests such as make a customer feedback survey or shorten my onboarding study.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fcreate-study\u002FSKILL.md",[],[203,168,204,205,30,206,170,207,173,208,175,209,210,177,211,212,213,214,215,180],"create study","create","brief","feedback","customer feedback","research brief","shorten","saved","research","requests","customer","questions","edit"]