{ "schemaVersion": "1.0", "item": { "slug": "hire", "name": "Hire", "source": "tencent", "type": "skill", "category": "ε…Άδ»–", "sourceUrl": "https://clawhub.ai/larsderidder/hire", "canonicalUrl": "https://clawhub.ai/larsderidder/hire", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/hire", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=hire", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md" ], "primaryDoc": "SKILL.md", "quickSetup": [ "Download the package from Yavira.", "Extract the archive and review SKILL.md first.", "Import or place the package into your OpenClaw setup." ], "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "sourceHealth": { "source": "tencent", "status": "healthy", "reason": "direct_download_ok", "recommendedAction": "download", "checkedAt": "2026-04-23T16:43:11.935Z", "expiresAt": "2026-04-30T16:43:11.935Z", "httpStatus": 200, "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentType": "application/zip", "probeMethod": "head", "details": { "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=4claw-imageboard", "contentDisposition": "attachment; filename=\"4claw-imageboard-1.0.1.zip\"", "redirectLocation": null, "bodySnippet": null }, "scope": "source", "summary": "Source download looks usable.", "detail": "Yavira can redirect you to the upstream package for this source.", "primaryActionLabel": "Download for OpenClaw", "primaryActionHref": "/downloads/hire" }, "validation": { "installChecklist": [ "Use the Yavira download entry.", "Review SKILL.md after the package is downloaded.", "Confirm the extracted package contains the expected setup assets." ], "postInstallChecks": [ "Confirm the extracted package includes the expected docs or setup files.", "Validate the skill or prompts are available in your target agent workspace.", "Capture any manual follow-up steps the agent could not complete." ] }, "downloadPageUrl": "https://openagent3.xyz/downloads/hire", "agentPageUrl": "https://openagent3.xyz/skills/hire/agent", "manifestUrl": "https://openagent3.xyz/skills/hire/agent.json", "briefUrl": "https://openagent3.xyz/skills/hire/agent.md" }, "agentAssist": { "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.", "steps": [ "Download the package from Yavira.", "Extract it into a folder your agent can access.", "Paste one of the prompts below and point your agent at the extracted folder." ], "prompts": [ { "label": "New install", "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Tell me what you changed and call out any manual steps you could not complete." }, { "label": "Upgrade existing", "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Summarize what changed and any follow-up checks I should run." } ] }, "documentation": { "source": "clawhub", "primaryDoc": "SKILL.md", "sections": [ { "title": "hire", "body": "Set up a new AI team member through a guided conversation. Not a config generator - a hiring process." }, { "title": "When to Use", "body": "User says something like:\n\n\"I want to hire a new agent\"\n\"I need help with X\" (where X implies a new agent role)\n\"Let's add someone to the team\"\n/hire" }, { "title": "3 core questions, asked one at a time:", "body": "Q1: \"What do you need help with?\"\nLet them describe the problem, not a job title. \"I'm drowning in code reviews\" beats \"I need a code reviewer.\"\n\nListen for: scope, implied autonomy level, implied tools needed\n\nQ2: \"What's their personality? Formal, casual, blunt, cautious, creative?\"\nOr frame it as: \"If this were a human colleague, what would they be like?\"\n\nListen for: communication style, vibe, how they interact\n\nQ3: \"What should they never do?\"\nThe red lines. This is where trust gets defined.\n\nListen for: boundaries, safety constraints, access limits" }, { "title": "Q4: Dynamic (optional)", "body": "After Q1-Q3, assess whether anything is ambiguous or needs clarification. If so, ask ONE follow-up question tailored to what's unclear. Examples:\n\n\"You mentioned monitoring - should they alert you immediately or batch updates?\"\n\"They'll need access to your codebase - any repos that are off-limits?\"\n\"You said 'casual' - are we talking friendly-professional or meme-level casual?\"\n\nIf Q1-Q3 were clear enough, skip Q4 entirely." }, { "title": "Summary Card", "body": "After the interview, present a summary:\n\n🎯 Role: [one-line description]\n🧠 Name: [suggested name from naming taxonomy]\nπŸ€– Model: [selected model] ([tier])\n⚑ Personality: [2-3 word vibe]\nπŸ”§ Tools: [inferred from conversation]\n🚫 Boundaries: [key red lines]\n🀝 Autonomy: [inferred level: high/medium/low]\n\nThen ask: \"Want to tweak anything, or are we good?\"" }, { "title": "Model Selection", "body": "Before finalizing, select an appropriate model for the agent." }, { "title": "Step 1: Discover available models", "body": "Run openclaw models list or check the gateway config to see what's configured." }, { "title": "Step 2: Categorize by tier", "body": "Map discovered models to capability tiers:\n\nTierModels (examples)Best forreasoningclaude-opus-, gpt-5, gpt-4o, deepseek-r1Strategy, advisory, complex analysis, architecturebalancedclaude-sonnet-*, gpt-4-turbo, gpt-4o-miniResearch, writing, general tasksfastclaude-haiku-, gpt-3.5, local/ollamaHigh volume, simple tasks, draftscodecodex-, claude-sonnet-, deepseek-coderCoding, refactoring, tests\n\nUse pattern matching on model names - don't hardcode specific versions." }, { "title": "Step 3: Match role to tier", "body": "Based on the interview:\n\nHeavy reasoning/advisory/strategy β†’ reasoning tier\nResearch/writing/creative β†’ balanced tier\nCode-focused β†’ code tier (or balanced if not available)\nHigh-volume/monitoring β†’ fast tier" }, { "title": "Step 4: Select and confirm", "body": "Pick the best available model for the role. In the summary card, add:\n\nπŸ€– Model: [selected model] ([tier] - [brief reason])\n\nIf multiple good options exist or you're unsure, ask:\n\"For a [role type] role, I'd suggest [model] (good balance of capability and cost). Or [alternative] if you want [deeper reasoning / faster responses / lower cost]. Preference?\"" }, { "title": "Notes", "body": "Don't assume any specific provider - work with what's available\nCheaper is better when capability is sufficient\nThe user's default model isn't always right for every agent\nIf only one model is available, use it and note it in the summary" }, { "title": "Optional Extras", "body": "After the summary is confirmed, offer:\n\n\"Want to set up periodic performance reviews?\"\n\nIf yes: ask preferred frequency (weekly, biweekly, monthly)\nCreate a cron job that triggers a review conversation\nReview covers: what went well, what's not working, scope/permission adjustments\nAt the end of each review, ask: \"Want to keep this schedule, change frequency, or stop reviews?\"\n\n\n\nOnboarding assignment (if relevant to the role)\n\nSuggest a small first task to test the new agent\nSomething real but low-stakes, so the user can see them in action" }, { "title": "What to Generate", "body": "Create an agent directory at agents// with:" }, { "title": "Always unique (generated fresh):", "body": "AGENTS.md - Role definition, responsibilities, operational rules, what they do freely vs ask first\nIDENTITY.md - Name, emoji, creature type, vibe, core principles" }, { "title": "Start from template, customize based on interview:", "body": "SOUL.md - Base from workspace SOUL.md template, customize vibe/boundaries sections\nTOOLS.md - Populated with inferred tools and access notes\nHEARTBEAT.md - Empty or with initial periodic tasks if relevant to role" }, { "title": "Symlink to shared (default, opinionated):", "body": "USER.md β†’ ../../USER.md (they need to know who they work for)\nMEMORY.md β†’ ../../MEMORY.md (shared team context)\n\nMention to the user: \"I've linked USER.md and MEMORY.md so they know who you are and share team context. You can change this later if you want them more isolated.\"" }, { "title": "Naming", "body": "Use craft/role-based names. Check TOOLS.md for the full naming taxonomy:\n\nResearch: Scout, Observer, Surveyor\nWriting: Scribe, Editor, Chronicler\nCode: Smith, Artisan, Engineer\nAnalysis: Analyst, Assessor, Arbiter\nCreative: Muse, Artisan\nOversight: Auditor, Reviewer, Warden\n\nCheck existing agents to avoid name conflicts. Suggest a name that fits the role, but let the user override." }, { "title": "Team Awareness", "body": "Before generating, check agents/ for existing team members. Note:\n\nPotential overlaps with existing roles\nGaps this new hire fills\nHow they'll interact with existing agents\n\nMention any relevant observations: \"You already have Scout for research - this new role would focus specifically on...\"" }, { "title": "After Setup", "body": "Tell the user what was created and where\n\n\nAutomatically update the OpenClaw config via gateway config.patch (do not ask the user to run a manual command). You must:\n\nAdd the new agent entry to agents.list using this format:\n{\n \"id\": \"\",\n \"workspace\": \"/home/lars/clawd/agents/\",\n \"model\": \"\"\n}\n\n\nAdd the new agent ID to the main agent's subagents.allowAgents array\nPreserve all existing agents and fields (arrays replace on patch)\n\nRequired flow:\n\nFetch config + hash\nopenclaw gateway call config.get --params '{}'\n\n\nBuild the updated agents.list array (existing entries + new agent) and update the main agent's subagents.allowAgents (existing list + new id).\nApply with config.patch:\nopenclaw gateway call config.patch --params '{\n \"raw\": \"{\\n agents: {\\n list: [ /* full list with new agent + updated main allowAgents */ ]\\n }\\n}\\n\",\n \"baseHash\": \"\",\n \"restartDelayMs\": 1000\n}'\n\n\n\n\n\nIf monthly reviews were requested, confirm the cron schedule\n\n\nUpdate any team roster if one exists" }, { "title": "Important", "body": "This is a CONVERSATION, not a form. Be natural.\nInfer as much as possible from context. Don't ask what you can figure out.\nThe user might not know what they want exactly. Help them figure it out.\nKeep the whole process under 5 minutes for the simple case." } ], "body": "hire\n\nSet up a new AI team member through a guided conversation. Not a config generator - a hiring process.\n\nWhen to Use\n\nUser says something like:\n\n\"I want to hire a new agent\"\n\"I need help with X\" (where X implies a new agent role)\n\"Let's add someone to the team\"\n/hire\nThe Interview\n3 core questions, asked one at a time:\n\nQ1: \"What do you need help with?\" Let them describe the problem, not a job title. \"I'm drowning in code reviews\" beats \"I need a code reviewer.\"\n\nListen for: scope, implied autonomy level, implied tools needed\n\nQ2: \"What's their personality? Formal, casual, blunt, cautious, creative?\" Or frame it as: \"If this were a human colleague, what would they be like?\"\n\nListen for: communication style, vibe, how they interact\n\nQ3: \"What should they never do?\" The red lines. This is where trust gets defined.\n\nListen for: boundaries, safety constraints, access limits\nQ4: Dynamic (optional)\n\nAfter Q1-Q3, assess whether anything is ambiguous or needs clarification. If so, ask ONE follow-up question tailored to what's unclear. Examples:\n\n\"You mentioned monitoring - should they alert you immediately or batch updates?\"\n\"They'll need access to your codebase - any repos that are off-limits?\"\n\"You said 'casual' - are we talking friendly-professional or meme-level casual?\"\n\nIf Q1-Q3 were clear enough, skip Q4 entirely.\n\nSummary Card\n\nAfter the interview, present a summary:\n\n🎯 Role: [one-line description]\n🧠 Name: [suggested name from naming taxonomy]\nπŸ€– Model: [selected model] ([tier])\n⚑ Personality: [2-3 word vibe]\nπŸ”§ Tools: [inferred from conversation]\n🚫 Boundaries: [key red lines]\n🀝 Autonomy: [inferred level: high/medium/low]\n\n\nThen ask: \"Want to tweak anything, or are we good?\"\n\nModel Selection\n\nBefore finalizing, select an appropriate model for the agent.\n\nStep 1: Discover available models\n\nRun openclaw models list or check the gateway config to see what's configured.\n\nStep 2: Categorize by tier\n\nMap discovered models to capability tiers:\n\nTier\tModels (examples)\tBest for\nreasoning\tclaude-opus-, gpt-5, gpt-4o, deepseek-r1\tStrategy, advisory, complex analysis, architecture\nbalanced\tclaude-sonnet-*, gpt-4-turbo, gpt-4o-mini\tResearch, writing, general tasks\nfast\tclaude-haiku-, gpt-3.5, local/ollama\tHigh volume, simple tasks, drafts\ncode\tcodex-, claude-sonnet-, deepseek-coder\tCoding, refactoring, tests\n\nUse pattern matching on model names - don't hardcode specific versions.\n\nStep 3: Match role to tier\n\nBased on the interview:\n\nHeavy reasoning/advisory/strategy β†’ reasoning tier\nResearch/writing/creative β†’ balanced tier\nCode-focused β†’ code tier (or balanced if not available)\nHigh-volume/monitoring β†’ fast tier\nStep 4: Select and confirm\n\nPick the best available model for the role. In the summary card, add:\n\nπŸ€– Model: [selected model] ([tier] - [brief reason])\n\n\nIf multiple good options exist or you're unsure, ask: \"For a [role type] role, I'd suggest [model] (good balance of capability and cost). Or [alternative] if you want [deeper reasoning / faster responses / lower cost]. Preference?\"\n\nNotes\nDon't assume any specific provider - work with what's available\nCheaper is better when capability is sufficient\nThe user's default model isn't always right for every agent\nIf only one model is available, use it and note it in the summary\nOptional Extras\n\nAfter the summary is confirmed, offer:\n\n\"Want to set up periodic performance reviews?\"\n\nIf yes: ask preferred frequency (weekly, biweekly, monthly)\nCreate a cron job that triggers a review conversation\nReview covers: what went well, what's not working, scope/permission adjustments\nAt the end of each review, ask: \"Want to keep this schedule, change frequency, or stop reviews?\"\n\nOnboarding assignment (if relevant to the role)\n\nSuggest a small first task to test the new agent\nSomething real but low-stakes, so the user can see them in action\nWhat to Generate\n\nCreate an agent directory at agents// with:\n\nAlways unique (generated fresh):\nAGENTS.md - Role definition, responsibilities, operational rules, what they do freely vs ask first\nIDENTITY.md - Name, emoji, creature type, vibe, core principles\nStart from template, customize based on interview:\nSOUL.md - Base from workspace SOUL.md template, customize vibe/boundaries sections\nTOOLS.md - Populated with inferred tools and access notes\nHEARTBEAT.md - Empty or with initial periodic tasks if relevant to role\nSymlink to shared (default, opinionated):\nUSER.md β†’ ../../USER.md (they need to know who they work for)\nMEMORY.md β†’ ../../MEMORY.md (shared team context)\n\nMention to the user: \"I've linked USER.md and MEMORY.md so they know who you are and share team context. You can change this later if you want them more isolated.\"\n\nNaming\n\nUse craft/role-based names. Check TOOLS.md for the full naming taxonomy:\n\nResearch: Scout, Observer, Surveyor\nWriting: Scribe, Editor, Chronicler\nCode: Smith, Artisan, Engineer\nAnalysis: Analyst, Assessor, Arbiter\nCreative: Muse, Artisan\nOversight: Auditor, Reviewer, Warden\n\nCheck existing agents to avoid name conflicts. Suggest a name that fits the role, but let the user override.\n\nTeam Awareness\n\nBefore generating, check agents/ for existing team members. Note:\n\nPotential overlaps with existing roles\nGaps this new hire fills\nHow they'll interact with existing agents\n\nMention any relevant observations: \"You already have Scout for research - this new role would focus specifically on...\"\n\nAfter Setup\n\nTell the user what was created and where\n\nAutomatically update the OpenClaw config via gateway config.patch (do not ask the user to run a manual command). You must:\n\nAdd the new agent entry to agents.list using this format:\n{\n \"id\": \"\",\n \"workspace\": \"/home/lars/clawd/agents/\",\n \"model\": \"\"\n}\n\nAdd the new agent ID to the main agent's subagents.allowAgents array\nPreserve all existing agents and fields (arrays replace on patch)\n\nRequired flow:\n\nFetch config + hash\nopenclaw gateway call config.get --params '{}'\n\nBuild the updated agents.list array (existing entries + new agent) and update the main agent's subagents.allowAgents (existing list + new id).\nApply with config.patch:\nopenclaw gateway call config.patch --params '{\n \"raw\": \"{\\n agents: {\\n list: [ /* full list with new agent + updated main allowAgents */ ]\\n }\\n}\\n\",\n \"baseHash\": \"\",\n \"restartDelayMs\": 1000\n}'\n\n\nIf monthly reviews were requested, confirm the cron schedule\n\nUpdate any team roster if one exists\n\nImportant\nThis is a CONVERSATION, not a form. Be natural.\nInfer as much as possible from context. Don't ask what you can figure out.\nThe user might not know what they want exactly. Help them figure it out.\nKeep the whole process under 5 minutes for the simple case." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/larsderidder/hire", "publisherUrl": "https://clawhub.ai/larsderidder/hire", "owner": "larsderidder", "version": "1.0.1", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/hire", "downloadUrl": "https://openagent3.xyz/downloads/hire", "agentUrl": "https://openagent3.xyz/skills/hire/agent", "manifestUrl": "https://openagent3.xyz/skills/hire/agent.json", "briefUrl": "https://openagent3.xyz/skills/hire/agent.md" } }