[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M76XVTQDR9EMEGE2KYC3HT1S":3,"skill-same:01M76XVTQDR9EMEGE2KYC3HT1S":47},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":24,"commitSha":25,"skillDirectory":26,"forksCount":27,"repositoryHtmlUrl":28,"repositoryHomepageUrl":29,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":31,"updatedAt":32,"install":33,"resources":46},"01M76XVTQDR9EMEGE2KYC3HT1S","agent-builder-plus",null,"Build high-performing OpenClaw agents end-to-end with comprehensive safety features. Use when you want to design a new agent (persona + operating rules) and generate required OpenClaw workspace files (SOUL.md, IDENTITY.md, AGENTS.md, USER.md, HEARTBEAT.md, optional MEMORY.md + memory\u002FYYYY-MM-DD.md). Includes anti-deadlock protection, timeout handling, error recovery, loop breaker, message overload protection, token limit protection, retry mechanism, health check, degraded mode, monitoring & logging, configuration validation, and channel conflict prevention. Also use to iterate on an existing agent's behavior, guardrails, autonomy model, heartbeat plan, and skill roster.","Lord1Egypt\u002Fawesome-skill-forge","Lord1Egypt","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F99746359?v=4","01M7592YDNT0XK9WJ0T2MQVPDN","Mohamed Mounir Ali AL Keshita","01M75GRFNHC8VZNWEB8DKDE433",3,"MIT","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fagent-builder-plus\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: agent-builder-plus\ndescription: Build high-performing OpenClaw agents end-to-end with comprehensive safety features. Use when you want to design a new agent (persona + operating rules) and generate required OpenClaw workspace files (SOUL.md, IDENTITY.md, AGENTS.md, USER.md, HEARTBEAT.md, optional MEMORY.md + memory\u002FYYYY-MM-DD.md). Includes anti-deadlock protection, timeout handling, error recovery, loop breaker, message overload protection, token limit protection, retry mechanism, health check, degraded mode, monitoring & logging, configuration validation, and channel conflict prevention. Also use to iterate on an existing agent's behavior, guardrails, autonomy model, heartbeat plan, and skill roster.\n---\n\n# Agent Builder Plus (OpenClaw)\n\nDesign and generate a complete **OpenClaw agent workspace** with strong defaults and advanced-user-oriented clarifying questions.\n\n## Quick Start\n\n```bash\n# 1. Read skill documentation\nRead SKILL.md and references\u002Fopenclaw-workspace.md\n\n# 2. Answer interview questions\nProvide answers for: job statement, surfaces, autonomy level, prohibitions, memory, tone, tool posture\n\n# 3. Generate workspace files\nThe skill will create: IDENTITY.md, SOUL.md, AGENTS.md, USER.md, HEARTBEAT.md\n\n# 4. Verify and test\nRun acceptance tests to validate behavior\n```\n\n## Canonical references\n\n- Workspace layout + heartbeat rules: **Read** `references\u002Fopenclaw-workspace.md`\n- File templates\u002Fsnippets: **Read** `references\u002Ftemplates.md`\n- Optional background (generic agent architecture): `references\u002Farchitecture.md`\n\n## Workflow: build an agent from scratch\n\n### Phase 1 - Interview (ask clarifying questions)\n\nAsk only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.\n\nMinimum question set (advanced):\n\n1) **Job statement**: What is the agent's primary mission in one sentence?\n2) **Surfaces**: Which channels (Telegram\u002FWhatsApp\u002FDiscord\u002FiMessage\u002FFeishu)? DM only vs groups?\n3) **Autonomy level**:\n   - Advisor (suggest only)\n   - Operator (non-destructive ok; ask before destructive\u002Fexternal)\n   - Autopilot (broad autonomy; higher risk)\n4) **Hard prohibitions**: Any actions the agent must never take?\n5) **Memory**: Should it keep curated `MEMORY.md`? What categories matter?\n6) **Tone**: concise vs narrative; strict vs warm; profanity rules; \"not the user's voice\" in groups?\n7) **Tool posture**: tool-first vs answer-first; verification requirements.\n\n**Error recovery:**\n\n- If user provides incomplete answers: Ask follow-up questions for missing information\n- If user is unsure about autonomy level: Explain trade-offs and suggest starting with \"Operator\"\n- If user wants to skip questions: Explain why each question matters for agent behavior\n\n### Phase 2 - Generate workspace files\n\nGenerate these files (minimum viable OpenClaw agent):\n\n- `IDENTITY.md`\n- `SOUL.md`\n- `AGENTS.md`\n- `USER.md`\n- `HEARTBEAT.md` (often empty by default)\n- `BOOTSTRAP.md` (for first-run guidance)\n\nOptionals:\n\n- `MEMORY.md` (private sessions only)\n- `memory\u002FYYYY-MM-DD.md` seed (today) with a short \"agent created\" entry\n- `TOOLS.md` starter (if the user wants per-environment notes)\n\n**File creation commands:**\n\n```bash\n# Create workspace directory\nmkdir -p \u002Fpath\u002Fto\u002Fworkspace\u002Fmemory\n\n# Create files (use write tool with correct parameters)\n# Important: Use `file_path` parameter, not `path`\n# Example:\n# write:\n#   file_path: \u002Fpath\u002Fto\u002Fworkspace\u002FIDENTITY.md\n#   content: \"content here\"\n\n# For large files (>2000 bytes), consider using file-writer skill\n# or split content into smaller chunks\n```\n\n**Note:** If you encounter \"Missing required parameter: path (path or file_path)\" error,\nensure you're using `file_path` parameter in your write tool calls.\n\n**Error handling:**\n\n- If directory creation fails: Check permissions and path validity\n- If file write fails: Verify disk space and write permissions\n- If template reference fails: Ensure references\u002Ftemplates.md exists\n\n**Error recovery:**\n\n- If directory creation fails: Check parent directory permissions, try alternative path\n- If file write fails: Check disk space, verify write permissions, retry with reduced content\n- If template reference fails: Verify references\u002F directory exists, check file permissions\n\nUse templates from `references\u002Ftemplates.md` but tailor content to the answers.\n\n### Phase 2.5 - Register agent with OpenClaw\n\n**⚠️ CRITICAL WARNING: Channel Conflict Prevention**\n\n**NEVER bind a new agent to the same channel as the main agent!**\n\nThis will cause the new agent to hijack the main agent's channel, making it impossible to communicate with the main agent.\n\n**Before registering, check existing agent bindings:**\n\n```bash\n# List all agents and their bindings\nopenclaw agents list\n\n# Check which channels are already in use\nopenclaw channels list\n```\n\n**Channel binding rules:**\n\n1. **Main agent (大鱼)**: Always uses the primary Feishu DM channel\n2. **New agents**: Must use DIFFERENT channels or sub-channels\n3. **Testing**: Use `\u002Fagentname` command binding for testing\n4. **Production**: Create separate Feishu apps or use different channels\n\n**Backup configuration first:**\n\n```bash\n# Backup openclaw.json before modification\ncp ~\u002F.openclaw\u002Fopenclaw.json ~\u002F.openclaw\u002Fopenclaw.json.backup\n```\n\n**Register the agent:**\n\n```bash\n# Add agent to OpenClaw configuration\nopenclaw agents add \u003Cagent-name> --workspace \u002Fpath\u002Fto\u002Fworkspace\n\n# Example 1: Use independent workspace (recommended - each agent has its own workspace)\nopenclaw agents add my-assistant --workspace ~\u002F.openclaw\u002Fworkspace-my-assistant\n\n# Example 2: Use default workspace (for single agent setup only)\nopenclaw agents add my-assistant --workspace ~\u002F.openclaw\u002Fworkspace\n```\n\n**Channel binding options:**\n\n**Feishu:**\n- **Direct binding:** Configure in Feishu app settings (recommended for production)\n  - ⚠️ WARNING: Do NOT bind to the same Feishu app as main agent\n  - Create a separate Feishu app for the new agent\n- **Command binding:** Use `\u002Fagentname` in Feishu messages (for testing)\n  - This is SAFE - does not hijack channels\n- **Multiple channels:** The agent can be bound to multiple channels simultaneously\n\n**Telegram:**\n- Create a separate bot token for each agent\n- Do NOT share bot tokens between agents\n- Use different bot usernames\n\n**WhatsApp:**\n- Use different phone numbers for each agent\n- Do NOT share WhatsApp Business API credentials\n\n**Discord:**\n- Use different bot tokens for each agent\n- Create separate Discord applications\n- Do NOT share bot tokens\n\n**iMessage:**\n- Each agent should use a different Apple ID\n- Do NOT share iMessage credentials\n\n**Authentication configuration (if needed):**\n\n```bash\n# Edit auth-profiles.json for external service access\n# Location: ~\u002F.openclaw\u002Fauth-profiles.json\n\n# Example structure:\n{\n  \"feishu\": {\n    \"appId\": \"cli_xxxxx\",\n    \"appSecret\": \"xxxxx\"\n  },\n  \"telegram\": {\n    \"botToken\": \"your-bot-token\"\n  }\n}\n```\n\n**Model provider configuration (optional):**\n\nIf the new agent needs to use custom model providers, you need to configure API keys:\n\n**Step 1: Check main agent's model configuration**\n\n```bash\n# View main agent's model configuration\ncat ~\u002F.openclaw\u002Fagents\u002Fmain\u002Fagent\u002Fmodels.json\n\n# View global model providers\ncat ~\u002F.openclaw\u002Fopenclaw.json | grep -A 20 \"providers\"\n```\n\n**Step 2: Choose configuration approach**\n\n**Option A: Use same model as main agent (recommended)**\n- New agent can directly use main agent's model configuration\n- No additional configuration needed\n- Models are shared across agents by default\n\n**Option B: Configure new model provider**\n- Required only if agent needs different model provider\n- Add provider to openclaw.json models.providers\n- Configure auth-profiles.json for the agent\n\n**Step 3: Configure model provider (if needed)**\n\n```bash\n# Add provider to openclaw.json\n# Location: ~\u002F.openclaw\u002Fopenclaw.json\n\n# Example structure:\n{\n  \"models\": {\n    \"providers\": {\n      \"custom-provider\": {\n        \"baseUrl\": \"https:\u002F\u002Fapi.example.com\u002Fv1\",\n        \"apiKey\": \"key_id:secret\",\n        \"api\": \"openai-completions\",\n        \"models\": [...]\n      }\n    }\n  }\n}\n```\n\n**Step 4: Configure agent-specific auth (if needed)**\n\n```bash\n# Edit auth-profiles.json for the agent\n# Location: ~\u002F.openclaw\u002Fagents\u002F\u003Cagent-name>\u002Fagent\u002Fauth-profiles.json\n\n# Example structure:\n{\n  \"custom-provider\": {\n    \"apiKey\": \"key_id:secret\",\n    \"baseUrl\": \"https:\u002F\u002Fapi.example.com\u002Fv1\"\n  }\n}\n```\n\n**⚠️ IMPORTANT: Model providers vs Channel providers**\n\n- **Model providers**: Configure AI model API keys (OpenAI, Anthropic, custom providers)\n  - Location: openclaw.json → models.providers\n  - Used for: AI model access\n  - Examples: openai, anthropic, custom-maas-api-*\n\n- **Channel providers**: Configure messaging platform credentials (Feishu, Telegram, Discord)\n  - Location: openclaw.json → channels.\u003Cprovider>\n  - Used for: Message delivery\n  - Examples: feishu, telegram, discord, whatsapp\n\n**Do not confuse these two types of providers!**\n\n**Common error:**\n```\n⚠️ Agent failed before reply: No API key found for provider \"provider-name\"\n```\n\n**Solution:**\n1. Check if provider name matches openclaw.json models.providers\n2. Verify API key is configured correctly\n3. Check auth-profiles.json for agent-specific configuration\n4. Restart OpenClaw Gateway after configuration changes\n\n**⚠️ IMPORTANT: Never reuse credentials from main agent!**\n\n**Error recovery:**\n\n- If `openclaw agents add` fails: Restores from backup and checks syntax\n- If auth fails: Verify credentials in auth-profiles.json\n- If binding fails: Check channel permissions and network connectivity\n- If backup fails: Check write permissions on ~\u002F.openclaw\u002F directory\n- **If main agent stops responding:** Immediately restore from backup and restart OpenClaw\n\n**Common errors and solutions:**\n\n**Error 1: \"No API key found for provider 'provider-name'\"**\n\n**Causes:**\n- Provider name doesn't match any configured provider in openclaw.json\n- Typo in provider name\n- Provider not added to models.providers\n\n**Solutions:**\n1. Check provider name in openclaw.json:\n   ```bash\n   cat ~\u002F.openclaw\u002Fopenclaw.json | grep -A 10 \"providers\"\n   ```\n2. Verify provider name spelling (case-sensitive)\n3. Add provider to models.providers if needed\n4. Restart OpenClaw Gateway after configuration changes\n\n**Error 2: \"Agent failed before reply\"**\n\n**Causes:**\n- Agent configuration is invalid\n- Workspace files are missing or corrupted\n- Model configuration is incorrect\n\n**Solutions:**\n1. Verify workspace files exist:\n   ```bash\n   ls -la \u002Fpath\u002Fto\u002Fworkspace\u002F\n   ```\n2. Check agent configuration:\n   ```bash\n   openclaw agents list\n   ```\n3. Test agent loading:\n   ```bash\n   openclaw agents test \u003Cagent-name>\n   ```\n4. Check OpenClaw logs:\n   ```bash\n   openclaw logs --follow\n   ```\n\n**Error 3: \"Missing required parameter: path (path or file_path)\"**\n\n**Causes:**\n- Using wrong parameter name in write tool\n- Parameter format error in tool call\n\n**Solutions:**\n1. Use correct parameter name: `file_path` instead of `path`\n2. Verify tool call syntax:\n   ```markdown\n   write:\n     file_path: \u002Fpath\u002Fto\u002Ffile.md\n     content: \"content here\"\n   ```\n3. Check tool documentation for correct parameters\n\n**Error 4: \"Could not find exact text in file\"**\n\n**Causes:**\n- Using edit tool with incorrect oldText\n- Whitespace differences (spaces vs tabs)\n- File already modified by another process\n\n**Solutions:**\n1. Re-read file to get exact content:\n   ```bash\n   read \u002Fpath\u002Fto\u002Ffile.md\n   ```\n2. Use unique markers for large sections:\n   ```markdown\n   \u003C!-- SECTION_START -->\n   [content]\n   \u003C!-- SECTION_END -->\n   ```\n3. Match whitespace exactly (copy from file read)\n4. Use smaller oldText for more precise matching\n\n### Phase 2.6 - Verify configuration\n\n**Verification steps:**\n\n```bash\n# 1. Check agent is registered\nopenclaw agents list\n\n# 2. Verify workspace files exist\nls -la \u002Fpath\u002Fto\u002Fworkspace\u002F\n\n# 3. Test agent can load (dry run)\nopenclaw agents test \u003Cagent-name>\n```\n\n**Success criteria:**\n\n- Agent appears in `openclaw agents list` output\n- All required files exist (IDENTITY.md, SOUL.md, AGENTS.md, USER.md)\n- No syntax errors in configuration files\n- Agent can be loaded without errors\n\n**If verification fails:**\n\n1. Check file permissions\n2. Validate JSON\u002FYAML syntax in configuration files\n3. Review error messages from `openclaw agents test`\n4. Restore from backup if necessary\n\n**Error recovery:**\n\n- If agent not in list: Check openclaw.json syntax, re-run Phase 2.5\n- If files missing: Re-run Phase 2 with corrected paths\n- If load test fails: Check file syntax, verify template content matches OpenClaw specs\n\n### Phase 3 - Guardrails checklist\n\nEnsure the generated agent includes:\n\n- Explicit ask-before-destructive rule.\n- Explicit ask-before-outbound-messages rule.\n- Stop-on-CLI-usage-error rule.\n- Max-iteration \u002F loop breaker guidance.\n- Group chat etiquette.\n- Sub-agent note: essential rules live in `AGENTS.md`.\n\n**Error recovery:**\n\n- If guardrails are missing: Add them to AGENTS.md or SOUL.md\n- If guardrails are too restrictive: Ask user for clarification on desired autonomy level\n\n### Phase 4 - Acceptance tests (fast)\n\nProvide 5-10 short scenario prompts to validate behavior, e.g.:\n\n- \"Draft but do not send a message to X; ask me before sending.\"\n- \"Summarize current workspace status without revealing secrets.\"\n- \"You hit an unknown flag error; show how you recover using --help.\"\n- \"In a group chat, someone asks something generic; decide whether to respond.\"\n\n**Error recovery:**\n\n- If agent fails acceptance tests: Review guardrails, adjust autonomy level, verify template content\n- If tests are too strict: Adjust test scenarios to match intended behavior\n\n### Phase 8 - Automated testing (optional)\n\n**Automated test commands:**\n\n```bash\n# Run OpenClaw's built-in agent tests\nopenclaw agents test \u003Cagent-name> --verbose\n\n# Test workspace file syntax\nopenclaw validate workspace \u002Fpath\u002Fto\u002Fworkspace\n\n# Test configuration loading\nopenclaw config test\n```\n\n**Test script example:**\n\n```bash\n#!\u002Fbin\u002Fbash\n# test-agent.sh - Automated agent testing script\n\nAGENT_NAME=\"my-assistant\"\nWORKSPACE=\"\u002Fpath\u002Fto\u002Fworkspace\"\n\necho \"Testing agent: $AGENT_NAME\"\n\n# Test 1: File existence\necho \"Test 1: Checking required files...\"\nfor file in IDENTITY.md SOUL.md AGENTS.md USER.md HEARTBEAT.md; do\n  if [ ! -f \"$WORKSPACE\u002F$file\" ]; then\n    echo \"❌ Missing: $file\"\n    exit 1\n  fi\ndone\necho \"✅ All required files present\"\n\n# Test 2: Configuration syntax\necho \"Test 2: Validating configuration...\"\nopenclaw agents list | grep -q \"$AGENT_NAME\"\nif [ $? -eq 0 ]; then\n  echo \"✅ Agent registered correctly\"\nelse\n  echo \"❌ Agent not found in configuration\"\n  exit 1\nfi\n\n# Test 3: Agent loading\necho \"Test 3: Testing agent load...\"\nopenclaw agents test \"$AGENT_NAME\"\nif [ $? -eq 0 ]; then\n  echo \"✅ Agent loads successfully\"\nelse\n  echo \"❌ Agent failed to load\"\n  exit 1\nfi\n\necho \"🎉 All tests passed!\"\n```\n\n**Error recovery:**\n\n- If automated tests fail: Review error messages, check file syntax, verify configuration\n- If test script fails: Check script permissions, verify paths are correct\n\n### Optional: systemd service configuration\n\nFor production deployments, the agent can run as a systemd service:\n\n\n**Service file example:**\n\n```bash\n# Create systemd service file\nsudo tee \u002Fetc\u002Fsystemd\u002Fsystem\u002Fopenclaw-agent.service > \u002Fdev\u002Fnull \u003C\u003CEOF\n[Unit]\nDescription=OpenClaw Agent Service\nAfter=network.target\n\n[Service]\nType=simple\nUser=\u003Cyour-username>\nWorkingDirectory=\u002Fhome\u002F\u003Cyour-username>\u002F.openclaw\nExecStart=\u002Fusr\u002Fbin\u002Fnode \u002Fhome\u002F\u003Cyour-username>\u002F.npm-global\u002Flib\u002Fnode_modules\u002Fopenclaw\u002Fdist\u002Findex.js\nRestart=always\nRestartSec=10\n\n[Install]\nWantedBy=multi-user.target\n\n\n[Install]\nWantedBy=multi-user.target\nEOF\n```\n\n**Enable and start service:**\n\n```bash\n# Reload systemd configuration\nsudo systemctl daemon-reload\n\n# Enable service to start on boot\nsudo systemctl enable openclaw-agent.service\n\n# Start service now\nsudo systemctl start openclaw-agent.service\n\n# Check status\nsudo systemctl status openclaw-agent.service\n\n# View logs\nsudo journalctl -u openclaw-agent.service -f\n```\n\n**Error recovery:**\n\n- If service fails to start: Check journalctl logs, verify file paths, check user permissions\n- If service crashes: Check OpenClaw logs, verify agent configuration, restart service\n- If service wont enable: Check systemd configuration syntax, verify systemd is running\n\n## Workflow: iterate on an existing agent\n\nWhen improving an existing agent, ask:\n\n1) What are the top 3 failure modes you have seen? (loops, overreach, verbosity, etc.)\n2) What autonomy changes do you want?\n3) Any new safety boundaries?\n4) Any changes to heartbeat behavior?\n\nThen propose targeted diffs to:\n\n- `SOUL.md` (persona\u002Ftone\u002Fboundaries)\n- `AGENTS.md` (operating rules + memory + delegation)\n- `HEARTBEAT.md` (small checklist)\n\nKeep changes minimal and surgical.\n\n**Error recovery:**\n\n- If diff application fails: Check file permissions, verify file exists, retry with backup\n- If changes break agent behavior: Restore from git or backup, review changes\n- If user rejects changes: Ask for clarification, propose alternative modifications\n","# Agent Builder Plus (OpenClaw)\n\nDesign and generate a complete **OpenClaw agent workspace** with strong defaults and advanced-user-oriented clarifying questions.\n\n## Quick Start\n\n```bash\n# 1. Read skill documentation\nRead SKILL.md and references\u002Fopenclaw-workspace.md\n\n# 2. Answer interview questions\nProvide answers for: job statement, surfaces, autonomy level, prohibitions, memory, tone, tool posture\n\n# 3. Generate workspace files\nThe skill will create: IDENTITY.md, SOUL.md, AGENTS.md, USER.md, HEARTBEAT.md\n\n# 4. Verify and test\nRun acceptance tests to validate behavior\n```\n\n## Canonical references\n\n- Workspace layout + heartbeat rules: **Read** `references\u002Fopenclaw-workspace.md`\n- File templates\u002Fsnippets: **Read** `references\u002Ftemplates.md`\n- Optional background (generic agent architecture): `references\u002Farchitecture.md`\n\n## Workflow: build an agent from scratch\n\n### Phase 1 - Interview (ask clarifying questions)\n\nAsk only what you need; keep it tight. Prefer multiple short rounds over one giant questionnaire.\n\nMinimum question set (advanced):\n\n1) **Job statement**: What is the agent's primary mission in one sentence?\n2) **Surfaces**: Which channels (Telegram\u002FWhatsApp\u002FDiscord\u002FiMessage\u002FFeishu)? DM only vs groups?\n3) **Autonomy level**:\n   - Advisor (suggest only)\n   - Operator (non-destructive ok; ask before destructive\u002Fexternal)\n   - Autopilot (broad autonomy; higher risk)\n4) **Hard prohibitions**: Any actions the agent must never take?\n5) **Memory**: Should it keep curated `MEMORY.md`? What categories matter?\n6) **Tone**: concise vs narrative; strict vs warm; profanity rules; \"not the user's voice\" in groups?\n7) **Tool posture**: tool-first vs answer-first; verification requirements.\n\n**Error recovery:**\n\n- If user provides incomplete answers: Ask follow-up questions for missing information\n- If user is unsure about autonomy level: Explain trade-offs and suggest starting with \"Operator\"\n- If user wants to skip questions: Explain why each question matters for agent behavior\n\n### Phase 2 - Generate workspace files\n\nGenerate these files (minimum viable OpenClaw agent):\n\n- `IDENTITY.md`\n- `SOUL.md`\n- `AGENTS.md`\n- `USER.md`\n- `HEARTBEAT.md` (often empty by default)\n- `BOOTSTRAP.md` (for first-run guidance)\n\nOptionals:\n\n- `MEMORY.md` (private sessions only)\n- `memory\u002FYYYY-MM-DD.md` seed (today) with a short \"agent created\" entry\n- `TOOLS.md` starter (if the user wants per-environment notes)\n\n**File creation commands:**\n\n```bash\n# Create workspace directory\nmkdir -p \u002Fpath\u002Fto\u002Fworkspace\u002Fmemory\n\n# Create files (use write tool with correct parameters)\n# Important: Use `file_path` parameter, not `path`\n# Example:\n# write:\n#   file_path: \u002Fpath\u002Fto\u002Fworkspace\u002FIDENTITY.md\n#   content: \"content here\"\n\n# For large files (>2000 bytes), consider using file-writer skill\n# or split content into smaller chunks\n```\n\n**Note:** If you encounter \"Missing required parameter: path (path or file_path)\" error,\nensure you're using `file_path` parameter in your write tool calls.\n\n**Error handling:**\n\n- If directory creation fails: Check permissions and path validity\n- If file write fails: Verify disk space and write permissions\n- If template reference fails: Ensure references\u002Ftemplates.md exists\n\n**Error recovery:**\n\n- If directory creation fails: Check parent directory permissions, try alternative path\n- If file write fails: Check disk space, verify write permissions, retry with reduced content\n- If template reference fails: Verify references\u002F directory exists, check file permissions\n\nUse templates from `references\u002Ftemplates.md` but tailor content to the answers.\n\n### Phase 2.5 - Register agent with OpenClaw\n\n**⚠️ CRITICAL WARNING: Channel Conflict Prevention**\n\n**NEVER bind a new agent to the same channel as the main agent!**\n\nThis will cause the new agent to hijack the main agent's channel, making it impossible to communicate with the main agent.\n\n**Before registering, check existing agent bindings:**\n\n```bash\n# List all agents and their bindings\nopenclaw agents list\n\n# Check which channels are already in use\nopenclaw channels list\n```\n\n**Channel binding rules:**\n\n1. **Main agent (大鱼)**: Always uses the primary Feishu DM channel\n2. **New agents**: Must use DIFFERENT channels or sub-channels\n3. **Testing**: Use `\u002Fagentname` command binding for testing\n4. **Production**: Create separate Feishu apps or use different channels\n\n**Backup configuration first:**\n\n```bash\n# Backup openclaw.json before modification\ncp ~\u002F.openclaw\u002Fopenclaw.json ~\u002F.openclaw\u002Fopenclaw.json.backup\n```\n\n**Register the agent:**\n\n```bash\n# Add agent to OpenClaw configuration\nopenclaw agents add \u003Cagent-name> --workspace \u002Fpath\u002Fto\u002Fworkspace\n\n# Example 1: Use independent workspace (recommended - each agent has its own workspace)\nopenclaw agents add my-assistant --workspace ~\u002F.openclaw\u002Fworkspace-my-assistant\n\n# Example 2: Use default workspace (for single agent setup only)\nopenclaw agents add my-assistant --workspace ~\u002F.openclaw\u002Fworkspace\n```\n\n**Channel binding options:**\n\n**Feishu:**\n- **Direct binding:** Configure in Feishu app settings (recommended for production)\n  - ⚠️ WARNING: Do NOT bind to the same Feishu app as main agent\n  - Create a separate Feishu app for the new agent\n- **Command binding:** Use `\u002Fagentname` in Feishu messages (for testing)\n  - This is SAFE - does not hijack channels\n- **Multiple channels:** The agent can be bound to multiple channels simultaneously\n\n**Telegram:**\n- Create a separate bot token for each agent\n- Do NOT share bot tokens between agents\n- Use different bot usernames\n\n**WhatsApp:**\n- Use different phone numbers for each agent\n- Do NOT share WhatsApp Business API credentials\n\n**Discord:**\n- Use different bot tokens for each agent\n- Create separate Discord applications\n- Do NOT share bot tokens\n\n**iMessage:**\n- Each agent should use a different Apple ID\n- Do NOT share iMessage credentials\n\n**Authentication configuration (if needed):**\n\n```bash\n# Edit auth-profiles.json for external service access\n# Location: ~\u002F.openclaw\u002Fauth-profiles.json\n\n# Example structure:\n{\n  \"feishu\": {\n    \"appId\": \"cli_xxxxx\",\n    \"appSecret\": \"xxxxx\"\n  },\n  \"telegram\": {\n    \"botToken\": \"your-bot-token\"\n  }\n}\n```\n\n**Model provider configuration (optional):**\n\nIf the new agent needs to use custom model providers, you need to configure API keys:\n\n**Step 1: Check main agent's model configuration**\n\n```bash\n# View main agent's model configuration\ncat ~\u002F.openclaw\u002Fagents\u002Fmain\u002Fagent\u002Fmodels.json\n\n# View global model providers\ncat ~\u002F.openclaw\u002Fopenclaw.json | grep -A 20 \"providers\"\n```\n\n**Step 2: Choose configuration approach**\n\n**Option A: Use same model as main agent (recommended)**\n- New agent can directly use main agent's model configuration\n- No additional configuration needed\n- Models are shared across agents by default\n\n**Option B: Configure new model provider**\n- Required only if agent needs different model provider\n- Add provider to openclaw.json models.providers\n- Configure auth-profiles.json for the agent\n\n**Step 3: Configure model provider (if needed)**\n\n```bash\n# Add provider to openclaw.json\n# Location: ~\u002F.openclaw\u002Fopenclaw.json\n\n# Example structure:\n{\n  \"models\": {\n    \"providers\": {\n      \"custom-provider\": {\n        \"baseUrl\": \"https:\u002F\u002Fapi.example.com\u002Fv1\",\n        \"apiKey\": \"key_id:secret\",\n        \"api\": \"openai-completions\",\n        \"models\": [...]\n      }\n    }\n  }\n}\n```\n\n**Step 4: Configure agent-specific auth (if needed)**\n\n```bash\n# Edit auth-profiles.json for the agent\n# Location: ~\u002F.openclaw\u002Fagents\u002F\u003Cagent-name>\u002Fagent\u002Fauth-profiles.json\n\n# Example structure:\n{\n  \"custom-provider\": {\n    \"apiKey\": \"key_id:secret\",\n    \"baseUrl\": \"https:\u002F\u002Fapi.example.com\u002Fv1\"\n  }\n}\n```\n\n**⚠️ IMPORTANT: Model providers vs Channel providers**\n\n- **Model providers**: Configure AI model API keys (OpenAI, Anthropic, custom providers)\n  - Location: openclaw.json → models.providers\n  - Used for: AI model access\n  - Examples: openai, anthropic, custom-maas-api-*\n\n- **Channel providers**: Configure messaging platform credentials (Feishu, Telegram, Discord)\n  - Location: openclaw.json → channels.\u003Cprovider>\n  - Used for: Message delivery\n  - Examples: feishu, telegram, discord, whatsapp\n\n**Do not confuse these two types of providers!**\n\n**Common error:**\n```\n⚠️ Agent failed before reply: No API key found for provider \"provider-name\"\n```\n\n**Solution:**\n1. Check if provider name matches openclaw.json models.providers\n2. Verify API key is configured correctly\n3. Check auth-profiles.json for agent-specific configuration\n4. Restart OpenClaw Gateway after configuration changes\n\n**⚠️ IMPORTANT: Never reuse credentials from main agent!**\n\n**Error recovery:**\n\n- If `openclaw agents add` fails: Restores from backup and checks syntax\n- If auth fails: Verify credentials in auth-profiles.json\n- If binding fails: Check channel permissions and network connectivity\n- If backup fails: Check write permissions on ~\u002F.openclaw\u002F directory\n- **If main agent stops responding:** Immediately restore from backup and restart OpenClaw\n\n**Common errors and solutions:**\n\n**Error 1: \"No API key found for provider 'provider-name'\"**\n\n**Causes:**\n- Provider name doesn't match any configured provider in openclaw.json\n- Typo in provider name\n- Provider not added to models.providers\n\n**Solutions:**\n1. Check provider name in openclaw.json:\n   ```bash\n   cat ~\u002F.openclaw\u002Fopenclaw.json | grep -A 10 \"providers\"\n   ```\n2. Verify provider name spelling (case-sensitive)\n3. Add provider to models.providers if needed\n4. Restart OpenClaw Gateway after configuration changes\n\n**Error 2: \"Agent failed before reply\"**\n\n**Causes:**\n- Agent configuration is invalid\n- Workspace files are missing or corrupted\n- Model configuration is incorrect\n\n**Solutions:**\n1. Verify workspace files exist:\n   ```bash\n   ls -la \u002Fpath\u002Fto\u002Fworkspace\u002F\n   ```\n2. Check agent configuration:\n   ```bash\n   openclaw agents list\n   ```\n3. Test agent loading:\n   ```bash\n   openclaw agents test \u003Cagent-name>\n   ```\n4. Check OpenClaw logs:\n   ```bash\n   openclaw logs --follow\n   ```\n\n**Error 3: \"Missing required parameter: path (path or file_path)\"**\n\n**Causes:**\n- Using wrong parameter name in write tool\n- Parameter format error in tool call\n\n**Solutions:**\n1. Use correct parameter name: `file_path` instead of `path`\n2. Verify tool call syntax:\n   ```markdown\n   write:\n     file_path: \u002Fpath\u002Fto\u002Ffile.md\n     content: \"content here\"\n   ```\n3. Check tool documentation for correct parameters\n\n**Error 4: \"Could not find exact text in file\"**\n\n**Causes:**\n- Using edit tool with incorrect oldText\n- Whitespace differences (spaces vs tabs)\n- File already modified by another process\n\n**Solutions:**\n1. Re-read file to get exact content:\n   ```bash\n   read \u002Fpath\u002Fto\u002Ffile.md\n   ```\n2. Use unique markers for large sections:\n   ```markdown\n   \u003C!-- SECTION_START -->\n   [content]\n   \u003C!-- SECTION_END -->\n   ```\n3. Match whitespace exactly (copy from file read)\n4. Use smaller oldText for more precise matching\n\n### Phase 2.6 - Verify configuration\n\n**Verification steps:**\n\n```bash\n# 1. Check agent is registered\nopenclaw agents list\n\n# 2. Verify workspace files exist\nls -la \u002Fpath\u002Fto\u002Fworkspace\u002F\n\n# 3. Test agent can load (dry run)\nopenclaw agents test \u003Cagent-name>\n```\n\n**Success criteria:**\n\n- Agent appears in `openclaw agents list` output\n- All required files exist (IDENTITY.md, SOUL.md, AGENTS.md, USER.md)\n- No syntax errors in configuration files\n- Agent can be loaded without errors\n\n**If verification fails:**\n\n1. Check file permissions\n2. Validate JSON\u002FYAML syntax in configuration files\n3. Review error messages from `openclaw agents test`\n4. Restore from backup if necessary\n\n**Error recovery:**\n\n- If agent not in list: Check openclaw.json syntax, re-run Phase 2.5\n- If files missing: Re-run Phase 2 with corrected paths\n- If load test fails: Check file syntax, verify template content matches OpenClaw specs\n\n### Phase 3 - Guardrails checklist\n\nEnsure the generated agent includes:\n\n- Explicit ask-before-destructive rule.\n- Explicit ask-before-outbound-messages rule.\n- Stop-on-CLI-usage-error rule.\n- Max-iteration \u002F loop breaker guidance.\n- Group chat etiquette.\n- Sub-agent note: essential rules live in `AGENTS.md`.\n\n**Error recovery:**\n\n- If guardrails are missing: Add them to AGENTS.md or SOUL.md\n- If guardrails are too restrictive: Ask user for clarification on desired autonomy level\n\n### Phase 4 - Acceptance tests (fast)\n\nProvide 5-10 short scenario prompts to validate behavior, e.g.:\n\n- \"Draft but do not send a message to X; ask me before sending.\"\n- \"Summarize current workspace status without revealing secrets.\"\n- \"You hit an unknown flag error; show how you recover using --help.\"\n- \"In a group chat, someone asks something generic; decide whether to respond.\"\n\n**Error recovery:**\n\n- If agent fails acceptance tests: Review guardrails, adjust autonomy level, verify template content\n- If tests are too strict: Adjust test scenarios to match intended behavior\n\n### Phase 8 - Automated testing (optional)\n\n**Automated test commands:**\n\n```bash\n# Run OpenClaw's built-in agent tests\nopenclaw agents test \u003Cagent-name> --verbose\n\n# Test workspace file syntax\nopenclaw validate workspace \u002Fpath\u002Fto\u002Fworkspace\n\n# Test configuration loading\nopenclaw config test\n```\n\n**Test script example:**\n\n```bash\n#!\u002Fbin\u002Fbash\n# test-agent.sh - Automated agent testing script\n\nAGENT_NAME=\"my-assistant\"\nWORKSPACE=\"\u002Fpath\u002Fto\u002Fworkspace\"\n\necho \"Testing agent: $AGENT_NAME\"\n\n# Test 1: File existence\necho \"Test 1: Checking required files...\"\nfor file in IDENTITY.md SOUL.md AGENTS.md USER.md HEARTBEAT.md; do\n  if [ ! -f \"$WORKSPACE\u002F$file\" ]; then\n    echo \"❌ Missing: $file\"\n    exit 1\n  fi\ndone\necho \"✅ All required files present\"\n\n# Test 2: Configuration syntax\necho \"Test 2: Validating configuration...\"\nopenclaw agents list | grep -q \"$AGENT_NAME\"\nif [ $? -eq 0 ]; then\n  echo \"✅ Agent registered correctly\"\nelse\n  echo \"❌ Agent not found in configuration\"\n  exit 1\nfi\n\n# Test 3: Agent loading\necho \"Test 3: Testing agent load...\"\nopenclaw agents test \"$AGENT_NAME\"\nif [ $? -eq 0 ]; then\n  echo \"✅ Agent loads successfully\"\nelse\n  echo \"❌ Agent failed to load\"\n  exit 1\nfi\n\necho \"🎉 All tests passed!\"\n```\n\n**Error recovery:**\n\n- If automated tests fail: Review error messages, check file syntax, verify configuration\n- If test script fails: Check script permissions, verify paths are correct\n\n### Optional: systemd service configuration\n\nFor production deployments, the agent can run as a systemd service:\n\n\n**Service file example:**\n\n```bash\n# Create systemd service file\nsudo tee \u002Fetc\u002Fsystemd\u002Fsystem\u002Fopenclaw-agent.service > \u002Fdev\u002Fnull \u003C\u003CEOF\n[Unit]\nDescription=OpenClaw Agent Service\nAfter=network.target\n\n[Service]\nType=simple\nUser=\u003Cyour-username>\nWorkingDirectory=\u002Fhome\u002F\u003Cyour-username>\u002F.openclaw\nExecStart=\u002Fusr\u002Fbin\u002Fnode \u002Fhome\u002F\u003Cyour-username>\u002F.npm-global\u002Flib\u002Fnode_modules\u002Fopenclaw\u002Fdist\u002Findex.js\nRestart=always\nRestartSec=10\n\n[Install]\nWantedBy=multi-user.target\n\n\n[Install]\nWantedBy=multi-user.target\nEOF\n```\n\n**Enable and start service:**\n\n```bash\n# Reload systemd configuration\nsudo systemctl daemon-reload\n\n# Enable service to start on boot\nsudo systemctl enable openclaw-agent.service\n\n# Start service now\nsudo systemctl start openclaw-agent.service\n\n# Check status\nsudo systemctl status openclaw-agent.service\n\n# View logs\nsudo journalctl -u openclaw-agent.service -f\n```\n\n**Error recovery:**\n\n- If service fails to start: Check journalctl logs, verify file paths, check user permissions\n- If service crashes: Check OpenClaw logs, verify agent configuration, restart service\n- If service wont enable: Check systemd configuration syntax, verify systemd is running\n\n## Workflow: iterate on an existing agent\n\nWhen improving an existing agent, ask:\n\n1) What are the top 3 failure modes you have seen? (loops, overreach, verbosity, etc.)\n2) What autonomy changes do you want?\n3) Any new safety boundaries?\n4) Any changes to heartbeat behavior?\n\nThen propose targeted diffs to:\n\n- `SOUL.md` (persona\u002Ftone\u002Fboundaries)\n- `AGENTS.md` (operating rules + memory + delegation)\n- `HEARTBEAT.md` (small checklist)\n\nKeep changes minimal and surgical.\n\n**Error recovery:**\n\n- If diff application fails: Check file permissions, verify file exists, retry with backup\n- If changes break agent behavior: Restore from git or backup, review changes\n- If user rejects changes: Ask for clarification, propose alternative modifications\n",{"name":5,"description":7},"master","4aa004da461d820f4ea5499e58eb8d196cf99016","community\u002Fclawhub\u002Fa\u002Fagent-builder-plus",0,"https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge","https:\u002F\u002Fskill-explorer-rho.vercel.app","https:\u002F\u002Fgithub.com\u002FLord1Egypt","Unbreakable. Unstoppable. Unmatched⚡.\r\n Speaker of machine tongues 🤖  | World-builder of decentralized dreams 🌐 | Defender of the digital realm🛡️.","2026-06-10 08:21:01",{"source":8,"skillName":5,"commands":34,"prompt":44,"zipAvailable":45},[35,38,41],{"tool":36,"command":37},"npx","npx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-builder-plus",{"tool":39,"command":40},"bunx","bunx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-builder-plus",{"tool":42,"command":43},"pnpm","pnpm dlx skills add Lord1Egypt\u002Fawesome-skill-forge --skill agent-builder-plus","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fcommunity\u002Fclawhub\u002Fa\u002Fagent-builder-plus\u002FSKILL.md",true,[],[48,56,63,70,77,84],{"skillNo":49,"name":50,"nameZh":6,"description":51,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":52,"categories":53,"keywords":54,"publishedAt":55},"01M785WKJB99G3STR1NF9NRTJW","youtube-content","YouTube transcripts to summaries, threads, blogs.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fyoutube-content\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":55},"01M785WEFHB0G2S9YA6EPR3H0E","xurl","X\u002FTwitter via xurl CLI: post, search, DM, media, v2 API.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fxurl\u002FSKILL.md",[],[],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":55},"01M785W98EDN4EFJXJPMQEHCRC","touchdesigner-mcp","Control a running TouchDesigner instance via twozero MCP — create operators, set parameters, wire connections, execute Python, build real-time visuals. 36 native tools.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftouchdesigner-mcp\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":55},"01M785W5S9FP3QBH2V191E025P","test-driven-development","TDD: enforce RED-GREEN-REFACTOR, tests before code.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Ftest-driven-development\u002FSKILL.md",[],[],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":55},"01M785W2AS08NSSH9YATCV6FJB","teams-meeting-pipeline","Operate the Teams meeting summary pipeline via Hermes CLI — summarize meetings, inspect pipeline status, replay jobs, manage Microsoft Graph subscriptions.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fteams-meeting-pipeline\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":55},"01M785VXKW4T34BTP1SCHW1R7F","systematic-debugging","4-phase root cause debugging: understand bugs before fixing.","https:\u002F\u002Fgithub.com\u002FLord1Egypt\u002Fawesome-skill-forge\u002Fblob\u002Fmaster\u002Fskills\u002Fsystematic-debugging\u002FSKILL.md",[],[]]