{ "schemaVersion": "1.0", "item": { "slug": "claw-agent-creator-archit", "name": "OpenClaw Agent Creator", "source": "tencent", "type": "skill", "category": "效率提升", "sourceUrl": "https://clawhub.ai/arch1904/claw-agent-creator-archit", "canonicalUrl": "https://clawhub.ai/arch1904/claw-agent-creator-archit", "targetPlatform": "OpenClaw" }, "install": { "downloadMode": "redirect", "downloadUrl": "/downloads/claw-agent-creator-archit", "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=claw-agent-creator-archit", "sourcePlatform": "tencent", "targetPlatform": "OpenClaw", "installMethod": "Manual import", "extraction": "Extract archive", "prerequisites": [ "OpenClaw" ], "packageFormat": "ZIP package", "includedAssets": [ "SKILL.md", "assets/templates/AGENTS.md", "assets/templates/HEARTBEAT.md", "assets/templates/IDENTITY.md", "assets/templates/SOUL.md", "assets/templates/USER.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/claw-agent-creator-archit" }, "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/claw-agent-creator-archit", "agentPageUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/agent", "manifestUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/agent.json", "briefUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/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": "OpenClaw Agent Creator", "body": "Create and configure agents for Arch's OpenClaw multi-agent system at ~/.openclaw/." }, { "title": "System Context", "body": "Owner: Archit (Arch), Linux user archit, timezone America/Denver\nGateway: Single process on port 18789 managing all agents\nBot: One Telegram bot shared across all agents — routing determines which agent handles which chat\nExisting agents: Check ~/.openclaw/openclaw.json → agents.list[] for current roster\nImplementation history: See ~/.openclaw/implementation-docs/ for the Wire agent reference implementation" }, { "title": "1. Gather Requirements", "body": "Before creating anything, clarify with Arch:\n\nAgent name and ID (lowercase, no spaces for ID)\nRole and responsibilities (specific, not vague)\nModel tier: cheap (Kimi K2.5 only) or full cascade (include Claude Sonnet)\nWhether it needs a Telegram group for Q&A\nWhether it needs cron jobs (what schedule, what tasks)\nWhether heartbeat should be enabled or disabled" }, { "title": "2. Stop the Gateway", "body": "openclaw gateway stop\n\nMANDATORY before editing openclaw.json or cron/jobs.json. The gateway actively writes to jobs.json (updating job state after each cron run). Editing while the gateway runs causes race conditions and data loss." }, { "title": "3. Backup Config", "body": "cp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.backup.$(date +%Y%m%d%H%M%S)" }, { "title": "4. Create Directories", "body": "mkdir -p ~/.openclaw/workspace-/memory\nmkdir -p ~/.openclaw/agents//agent\n\nNEVER reuse agentDir across agents — causes auth/session collisions." }, { "title": "5. Write Workspace Files", "body": "Use templates from assets/templates/ as starting points. Every agent needs:\n\nFilePurposeRequiredSOUL.mdPersonality, role, responsibilities, behavioral modesYesIDENTITY.mdQuick-reference card (name, role, emoji)YesUSER.mdAbout Arch (copy from any existing agent workspace)YesAGENTS.mdWorkspace rules (boot sequence, memory, safety)YesHEARTBEAT.mdPeriodic task checklist (or comment if disabled)Yes\n\nSOUL.md is the most important file. Be specific about responsibilities. Include behavioral modes if the agent operates differently in different contexts (e.g., briefing mode vs chat mode)." }, { "title": "6. Edit openclaw.json — Agent Entry", "body": "Add to agents.list[]. See references/config-schema.md for all valid fields.\n\nMinimal entry:\n\n{\n \"id\": \"\",\n \"name\": \"\",\n \"workspace\": \"/home/archit/.openclaw/workspace-\",\n \"agentDir\": \"/home/archit/.openclaw/agents//agent\",\n \"identity\": { \"name\": \"\" }\n}\n\nCommon additions:\n\n\"model\" — Override the default model cascade. Exclude expensive models for worker agents.\n\"heartbeat\": { \"every\": \"0\" } — Disable heartbeat for cron-only agents.\n\"groupChat\": { \"mentionPatterns\": [\"@\", \"@\"] } — Enable @mentions in groups.\n\nOnly ONE agent should have \"default\": true (currently Fossil). The default agent receives all unrouted messages." }, { "title": "7. Edit openclaw.json — Telegram Routing (if needed)", "body": "THREE separate config changes are required. Missing any one causes silent failures. See references/telegram-routing.md for the full explanation.\n\nGroup config in channels.telegram.groups:\n\"-100XXXXXXXXXX\": { \"requireMention\": false }\n\n\n\nBinding in bindings[]:\n{ \"agentId\": \"\", \"match\": { \"channel\": \"telegram\", \"peer\": { \"kind\": \"group\", \"id\": \"-100XXXXXXXXXX\" } } }\n\n\n\nMention patterns on the agent entry (already done in step 6 if groupChat was added)." }, { "title": "8. Create Cron Jobs (if needed)", "body": "Edit cron/jobs.json. Every cron job prompt MUST include:\n\nDynamic group ID resolution preamble (NEVER hardcode Telegram group IDs):\nFIRST: Resolve your Telegram group ID by running:\njq -r '.bindings[] | select(.agentId == \"\") | .match.peer.id' ~/.openclaw/openclaw.json\nUse the output as the target for all Telegram messages in this task.\n\n\nDate injection: $(date '+%A, %B %d, %Y') after the preamble\nExplicit constraints: source allowlists, recency rules, format templates\nDelivery instructions: use target='' placeholder (resolved by the preamble)\n\nThis self-healing pattern ensures cron jobs survive Telegram group ID migrations. See references/prompt-patterns.md for full patterns and references/telegram-routing.md for why this matters.\n\nCritical: If copying files or prompts from another agent's workspace, grep for hardcoded paths and update them." }, { "title": "9. Restart Gateway and Verify", "body": "openclaw gateway start\n\nVerify in logs:\n\nAgent registered: agent registered: \nMessages route correctly: lane enqueue: lane=session:agent::...\n\nIf messages to a Telegram group show skip: no-mention, the channels.telegram.groups config is missing (see references/bugs-and-pitfalls.md)." }, { "title": "Reference Files", "body": "FileWhen to Readreferences/config-schema.mdWhen writing agent config or cron jobsreferences/telegram-routing.mdWhen setting up Telegram group routingreferences/prompt-patterns.mdWhen writing cron job promptsreferences/bugs-and-pitfalls.mdWhen debugging issues or before any config edit" }, { "title": "Template Files", "body": "Starter templates for workspace files are in assets/templates/. Copy and customize per agent." } ], "body": "OpenClaw Agent Creator\n\nCreate and configure agents for Arch's OpenClaw multi-agent system at ~/.openclaw/.\n\nSystem Context\nOwner: Archit (Arch), Linux user archit, timezone America/Denver\nGateway: Single process on port 18789 managing all agents\nBot: One Telegram bot shared across all agents — routing determines which agent handles which chat\nExisting agents: Check ~/.openclaw/openclaw.json → agents.list[] for current roster\nImplementation history: See ~/.openclaw/implementation-docs/ for the Wire agent reference implementation\nAgent Creation Workflow\n1. Gather Requirements\n\nBefore creating anything, clarify with Arch:\n\nAgent name and ID (lowercase, no spaces for ID)\nRole and responsibilities (specific, not vague)\nModel tier: cheap (Kimi K2.5 only) or full cascade (include Claude Sonnet)\nWhether it needs a Telegram group for Q&A\nWhether it needs cron jobs (what schedule, what tasks)\nWhether heartbeat should be enabled or disabled\n2. Stop the Gateway\nopenclaw gateway stop\n\n\nMANDATORY before editing openclaw.json or cron/jobs.json. The gateway actively writes to jobs.json (updating job state after each cron run). Editing while the gateway runs causes race conditions and data loss.\n\n3. Backup Config\ncp ~/.openclaw/openclaw.json ~/.openclaw/openclaw.json.backup.$(date +%Y%m%d%H%M%S)\n\n4. Create Directories\nmkdir -p ~/.openclaw/workspace-/memory\nmkdir -p ~/.openclaw/agents//agent\n\n\nNEVER reuse agentDir across agents — causes auth/session collisions.\n\n5. Write Workspace Files\n\nUse templates from assets/templates/ as starting points. Every agent needs:\n\nFile\tPurpose\tRequired\nSOUL.md\tPersonality, role, responsibilities, behavioral modes\tYes\nIDENTITY.md\tQuick-reference card (name, role, emoji)\tYes\nUSER.md\tAbout Arch (copy from any existing agent workspace)\tYes\nAGENTS.md\tWorkspace rules (boot sequence, memory, safety)\tYes\nHEARTBEAT.md\tPeriodic task checklist (or comment if disabled)\tYes\n\nSOUL.md is the most important file. Be specific about responsibilities. Include behavioral modes if the agent operates differently in different contexts (e.g., briefing mode vs chat mode).\n\n6. Edit openclaw.json — Agent Entry\n\nAdd to agents.list[]. See references/config-schema.md for all valid fields.\n\nMinimal entry:\n\n{\n \"id\": \"\",\n \"name\": \"\",\n \"workspace\": \"/home/archit/.openclaw/workspace-\",\n \"agentDir\": \"/home/archit/.openclaw/agents//agent\",\n \"identity\": { \"name\": \"\" }\n}\n\n\nCommon additions:\n\n\"model\" — Override the default model cascade. Exclude expensive models for worker agents.\n\"heartbeat\": { \"every\": \"0\" } — Disable heartbeat for cron-only agents.\n\"groupChat\": { \"mentionPatterns\": [\"@\", \"@\"] } — Enable @mentions in groups.\n\nOnly ONE agent should have \"default\": true (currently Fossil). The default agent receives all unrouted messages.\n\n7. Edit openclaw.json — Telegram Routing (if needed)\n\nTHREE separate config changes are required. Missing any one causes silent failures. See references/telegram-routing.md for the full explanation.\n\nGroup config in channels.telegram.groups:\n\n\"-100XXXXXXXXXX\": { \"requireMention\": false }\n\n\nBinding in bindings[]:\n\n{ \"agentId\": \"\", \"match\": { \"channel\": \"telegram\", \"peer\": { \"kind\": \"group\", \"id\": \"-100XXXXXXXXXX\" } } }\n\n\nMention patterns on the agent entry (already done in step 6 if groupChat was added).\n\n8. Create Cron Jobs (if needed)\n\nEdit cron/jobs.json. Every cron job prompt MUST include:\n\nDynamic group ID resolution preamble (NEVER hardcode Telegram group IDs):\nFIRST: Resolve your Telegram group ID by running:\njq -r '.bindings[] | select(.agentId == \"\") | .match.peer.id' ~/.openclaw/openclaw.json\nUse the output as the target for all Telegram messages in this task.\n\nDate injection: $(date '+%A, %B %d, %Y') after the preamble\nExplicit constraints: source allowlists, recency rules, format templates\nDelivery instructions: use target='' placeholder (resolved by the preamble)\n\nThis self-healing pattern ensures cron jobs survive Telegram group ID migrations. See references/prompt-patterns.md for full patterns and references/telegram-routing.md for why this matters.\n\nCritical: If copying files or prompts from another agent's workspace, grep for hardcoded paths and update them.\n\n9. Restart Gateway and Verify\nopenclaw gateway start\n\n\nVerify in logs:\n\nAgent registered: agent registered: \nMessages route correctly: lane enqueue: lane=session:agent::...\n\nIf messages to a Telegram group show skip: no-mention, the channels.telegram.groups config is missing (see references/bugs-and-pitfalls.md).\n\nReference Files\nFile\tWhen to Read\nreferences/config-schema.md\tWhen writing agent config or cron jobs\nreferences/telegram-routing.md\tWhen setting up Telegram group routing\nreferences/prompt-patterns.md\tWhen writing cron job prompts\nreferences/bugs-and-pitfalls.md\tWhen debugging issues or before any config edit\nTemplate Files\n\nStarter templates for workspace files are in assets/templates/. Copy and customize per agent." }, "trust": { "sourceLabel": "tencent", "provenanceUrl": "https://clawhub.ai/arch1904/claw-agent-creator-archit", "publisherUrl": "https://clawhub.ai/arch1904/claw-agent-creator-archit", "owner": "arch1904", "version": "1.0.0", "license": null, "verificationStatus": "Indexed source record" }, "links": { "detailUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit", "downloadUrl": "https://openagent3.xyz/downloads/claw-agent-creator-archit", "agentUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/agent", "manifestUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/agent.json", "briefUrl": "https://openagent3.xyz/skills/claw-agent-creator-archit/agent.md" } }