[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6CT7MP9QX0TS8W13PRGDWE2":3,"skill-same:01M6CT7MP9QX0TS8W13PRGDWE2":53},{"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":25,"linkOnly":26,"markdownContent":27,"markdownBody":28,"frontMatter":29,"refName":32,"commitSha":33,"skillDirectory":34,"forksCount":35,"repositoryHtmlUrl":36,"repositoryHomepageUrl":6,"ownerHtmlUrl":37,"ownerBlogUrl":6,"ownerDescription":38,"updatedAt":39,"install":40,"resources":52},"01M6CT7MP9QX0TS8W13PRGDWE2","customer-conversations",null,"Prepares briefings before customer calls and records what you learned afterward. Use before and after customer interviews or sales calls.","vibbs\u002Fcompany-os","vibbs","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F7379149?v=4","01M6CT6M03CAS1H7RA89F3965B","Vibbs Dod","01M6CT6M4MY60C9FFQ3AAR2RSK",6,"MIT","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcustomer-conversations\u002FSKILL.md",[],[19,20,21,22,23,24],"customer","customer conversations","before and after","sales","use","patterns","2026-11-10 10:12:30",false,"---\nname: customer-conversations\ndescription: Prepares briefings before customer calls and records what you learned afterward. Use before and after customer interviews or sales calls.\nuser-invocable: true\nargument-hint: \"[prep \u003Ctype> | debrief | patterns]\"\n---\n\n# Customer Conversations\n\n## Reference\n- **ID**: S-PROD-08\n- **Category**: Product\n- **Inputs**: company.config.yaml, existing CONV- artifacts, ICP artifacts, PRDs\n- **Outputs**: conversation log artifacts → artifacts\u002Fdecision-memos\u002FCONV-{date}-{type}-{company}.md\n- **Used by**: Product Agent (primary), Ops & Risk Agent (churn signals)\n- **Tool scripts**: .\u002Ftools\u002Fartifact\u002Fvalidate.sh\n\n## Purpose\n\nStructures the process of talking to customers. Provides pre-call preparation (what to ask, what to listen for), captures post-call debriefs in a structured format, and routes insights to the right downstream skills (ICP refinement, PRD evidence, churn detection).\n\n## When to Use\n\n- Before a customer call, demo, or interview — run `prep \u003Ctype>` for a tailored briefing\n- After any customer interaction — run `debrief` to capture structured insights\n- When you have 3+ conversation logs — run `patterns` to surface cross-conversation themes\n\n## Conversation Types\n\n| Type | Purpose | Key Signals |\n|------|---------|-------------|\n| `discovery` | Understand user problems, validate hypotheses | Pain intensity, current workarounds, willingness to pay |\n| `mom-test` | Mom-Test methodology — validate pain without leading | Current behavior, $$ cost of status quo, workarounds tried |\n| `demo` | Show the product, gauge interest | Feature reactions, objections, comparison mentions |\n| `churn-save` | Understand why a user is leaving | Root cause, what would change their mind, competitor mentions |\n| `upsell` | Explore expansion opportunities | Usage patterns, feature gaps, budget signals |\n| `check-in` | Relationship maintenance, satisfaction pulse | NPS proxy, feature requests, referral willingness |\n| `cold-outreach` | Reaching out to potential customers who haven't expressed interest yet | Pain points, willingness to pay signals, competitive mentions |\n\n## Procedure\n\n### Cold Outreach Conversation Type\n\n**Cold Outreach** conversation type:\n- **When**: Reaching out to potential customers who haven't expressed interest yet\n- **Prep**: Research the prospect, prepare problem-led opening, have 3 discovery questions ready\n- **During**: Listen 80%, talk 20%. Goal is learning, not selling.\n- **Debrief**: Capture pain points, willingness to pay signals, competitive mentions\n\n**Starting from Scratch Detection**: If no ICP artifact exists in `artifacts\u002Fproduct\u002F` and no CONV- artifacts exist in `artifacts\u002Fdecision-memos\u002F`, present a \"Starting from scratch\" briefing:\n- Guidance on finding first 10 targets (LinkedIn search, community lurking, competitor reviews)\n- Starter interview script focused on problem discovery\n- Link to icp-positioning skill to formalize learnings after 5+ conversations\n\n### Mom-Test Interview Methodology\n\nWhen conversation type is `mom-test`, apply The Mom Test rules strictly:\n\n**Core Principle**: Talk about their life, not your idea. Extract facts about past behavior, not opinions about the future.\n\n**The 5 Mom-Test Rules**:\n1. **Never ask \"Would you use this?\"** — People are polite liars. Instead ask: \"How do you handle [problem] today?\"\n2. **Never ask \"Would you pay for this?\"** — Instead ask: \"What does [current workaround] cost you in time\u002Fmoney?\"\n3. **Ask about specifics in the past, not hypotheticals** — \"Tell me about the last time [problem] happened\" vs. \"Do you ever experience [problem]?\"\n4. **Talk less, listen more** — Target 80% listening, 20% talking. Follow their energy.\n5. **Extract commitment or advancement** — End every call with a concrete next step (intro to a colleague, LOI review, pilot agreement).\n\n**Mom-Test Question Bank** (use 5-7 per call, 15 min max):\n- \"What's the hardest part of [workflow] for you right now?\"\n- \"Can you walk me through the last time that happened?\"\n- \"What did you do about it? What else have you tried?\"\n- \"How much time\u002Fmoney does that cost you per [week\u002Fmonth]?\"\n- \"Who else in your org deals with this problem?\"\n- \"What would a dream solution look like?\" (only AFTER understanding the problem)\n- \"If you had a magic wand, what would you change about [current tool\u002Fprocess]?\"\n- \"What are you using today to solve this? What's working? What's not?\"\n- \"Have you looked for solutions? What happened?\"\n- \"How urgent is solving this? What's your timeline?\"\n\n**Anti-Patterns to Avoid**:\n- Pitching your solution before minute 10\n- Asking binary yes\u002Fno questions\n- Fishing for compliments (\"Don't you think X would be great?\")\n- Talking about features before understanding the problem\n- Accepting \"that would be nice\" as validation (it's not)\n\n**Debrief Extras for Mom-Test**:\nWhen debriefing a mom-test conversation, also capture:\n- **$$ Quantification**: estimated cost of the problem per month\u002Fyear\n- **Workarounds tried**: what they've already done to solve this\n- **Commitment signal**: did they offer an intro, agree to review an LOI, or ask for a follow-up?\n\n### Mode: `prep \u003Ctype>`\n\n1. **Load Context**\n   - Read `company.config.yaml` for product name, stage, ICP\n   - Check `artifacts\u002Fproduct\u002F` for the most recent ICP positioning artifact\n   - Check `artifacts\u002Fprds\u002F` for in-progress features (gives context on what's being built)\n   - Check recent CONV- artifacts for patterns from this company\u002Fsegment\n\n2. **Generate Pre-Call Briefing**\n\n   Produce a briefing document with:\n\n   **Do Ask** (5 open-ended questions tailored to conversation type):\n   - Discovery: \"What's the hardest part of [workflow] for you right now?\"\n   - Demo: \"Which of these capabilities matters most to your day-to-day?\"\n   - Churn-save: \"What changed since you first signed up?\"\n   - Upsell: \"If you could wave a magic wand and add one capability, what would it be?\"\n   - Check-in: \"What's working well? What's frustrating?\"\n\n   **Don't Ask** (anti-patterns to avoid):\n   - Leading questions (\"Don't you think our feature X is great?\")\n   - Binary questions when you need depth (\"Do you like it?\" → \"What's your experience with it?\")\n   - Feature-first questions before understanding the problem\n\n   **Listen For** (signals specific to conversation type):\n   - ICP fit signals: role, company size, tech stack, budget authority\n   - Competitor mentions: who else they're evaluating or using\n   - Churn risk indicators: frustration level, timeline pressure, alternative mentions\n   - Willingness-to-pay signals: asking about pricing, comparing to current spend\n\n   **Context from Recent Conversations** (if CONV- artifacts exist for this company):\n   - Previous pain points mentioned\n   - Commitments made\n   - Open questions from last call\n\n3. **Present Briefing** — show the briefing to the user before the call. Do not save as an artifact (ephemeral output).\n\n### Mode: `debrief`\n\n4. **Capture Structured Debrief**\n\n   Prompt the user to provide (or extract from conversation notes):\n\n   | Field | Description |\n   |-------|-------------|\n   | **Date** | When the conversation happened |\n   | **Participant** | Name, role, company, segment |\n   | **Conversation type** | discovery \u002F demo \u002F churn-save \u002F upsell \u002F check-in |\n   | **Pain level** | 1 (minor inconvenience) to 5 (hair-on-fire) |\n   | **Key quotes** | 2-3 verbatim quotes that capture the essence |\n   | **Willingness to pay** | Strong signal \u002F Weak signal \u002F No signal \u002F Price sensitive |\n   | **Competitor mentions** | Any alternatives mentioned (by name) |\n   | **Churn signals** | None \u002F Low \u002F Medium \u002F High |\n   | **Upsell signals** | None \u002F Low \u002F Medium \u002F High |\n   | **Feature requests** | Specific capabilities mentioned |\n   | **ICP fit** | Strong \u002F Partial \u002F Weak — with reasoning |\n   | **Recommended next action** | Follow-up call, send proposal, route to support, etc. |\n\n5. **Classify and Tag**\n\n   Based on the debrief, assign tags:\n   - `prd_evidence: true` — if the conversation surfaced a clear product need\n   - `icp_signal: true` — if the conversation provided ICP refinement data\n   - `churn_signal: true` — if churn risk was detected (Medium or High)\n   - `upsell_signal: true` — if expansion opportunity was identified\n\n6. **Save Artifact**\n\n   Write to `artifacts\u002Fdecision-memos\u002FCONV-{date}-{type}-{company}.md`:\n\n   ```yaml\n   ---\n   id: CONV-{date}-{type}-{company}\n   type: decision-memo\n   title: \"{Type} Conversation — {Company}\"\n   status: review\n   created: {date}\n   tags: [customer-conversation, {type}]\n   prd_evidence: true|false\n   icp_signal: true|false\n   churn_signal: true|false\n   ---\n   ```\n\n   Body contains the structured debrief fields above.\n\n7. **Route Signals**\n\n   After saving:\n   - If `churn_signal: true` → inform the user: \"Churn signal detected. Consider running `\u002Fsupport-operations` to review escalation paths, or `\u002Factivation-onboarding` to design a re-engagement flow.\"\n   - If `icp_signal: true` → inform the user: \"ICP data captured. Run `\u002Ficp-positioning` to update your ideal customer profile with this evidence.\"\n   - If `prd_evidence: true` → inform the user: \"Product evidence captured. This will be consumed by `\u002Ffeedback-synthesizer` in the next synthesis cycle.\"\n\n8. **Validate** — run `.\u002Ftools\u002Fartifact\u002Fvalidate.sh` on the saved artifact.\n\n### Mode: `patterns`\n\n9. **Cross-Conversation Analysis**\n\n   Scan all CONV- artifacts in `artifacts\u002Fdecision-memos\u002F`:\n   - Count by conversation type\n   - Extract all `Key quotes` fields\n   - Tally pain levels by segment\n   - List all competitor mentions with frequency\n   - Identify recurring feature requests (3+ mentions = pattern)\n   - Aggregate churn signals by segment\n\n10. **Surface Patterns**\n\n    Present a summary:\n    - **Top pain points**: ranked by frequency × pain level\n    - **Competitor landscape**: who's mentioned most, in what context\n    - **ICP signals**: which segments show strongest fit\n    - **Emerging feature requests**: requests that appear 3+ times\n    - **Churn risk segments**: segments with concentrated churn signals\n\n    This output is ephemeral (presented to user, not saved as artifact). The user can then feed specific findings into `\u002Ffeedback-synthesizer` or `\u002Ficp-positioning`.\n\n## Quality Checklist\n- [ ] Pre-call briefings include 5 open-ended questions (no leading questions)\n- [ ] Debriefs capture all structured fields\n- [ ] Artifact frontmatter uses type `decision-memo` (passes validate.sh)\n- [ ] Signal routing recommendations are provided\n- [ ] Pattern analysis only runs with 3+ CONV- artifacts\n- [ ] Artifact passes validation\n\n## Cross-References\n- **feedback-synthesizer** (S-PROD-04) — consumes CONV- artifacts as input during synthesis\n- **icp-positioning** (S-PROD-01) — CONV- artifacts provide ICP evidence\n- **prd-writer** (S-PROD-02) — CONV- artifacts populate PRD evidence sections\n- **support-operations** (S-OPS-07) — churn signals feed support escalation review\n- **inbound-loop-sop** — `standards\u002Fops\u002Finbound-loop-sop.md` documents the full pipeline\n","# Customer Conversations\n\n## Reference\n- **ID**: S-PROD-08\n- **Category**: Product\n- **Inputs**: company.config.yaml, existing CONV- artifacts, ICP artifacts, PRDs\n- **Outputs**: conversation log artifacts → artifacts\u002Fdecision-memos\u002FCONV-{date}-{type}-{company}.md\n- **Used by**: Product Agent (primary), Ops & Risk Agent (churn signals)\n- **Tool scripts**: .\u002Ftools\u002Fartifact\u002Fvalidate.sh\n\n## Purpose\n\nStructures the process of talking to customers. Provides pre-call preparation (what to ask, what to listen for), captures post-call debriefs in a structured format, and routes insights to the right downstream skills (ICP refinement, PRD evidence, churn detection).\n\n## When to Use\n\n- Before a customer call, demo, or interview — run `prep \u003Ctype>` for a tailored briefing\n- After any customer interaction — run `debrief` to capture structured insights\n- When you have 3+ conversation logs — run `patterns` to surface cross-conversation themes\n\n## Conversation Types\n\n| Type | Purpose | Key Signals |\n|------|---------|-------------|\n| `discovery` | Understand user problems, validate hypotheses | Pain intensity, current workarounds, willingness to pay |\n| `mom-test` | Mom-Test methodology — validate pain without leading | Current behavior, $$ cost of status quo, workarounds tried |\n| `demo` | Show the product, gauge interest | Feature reactions, objections, comparison mentions |\n| `churn-save` | Understand why a user is leaving | Root cause, what would change their mind, competitor mentions |\n| `upsell` | Explore expansion opportunities | Usage patterns, feature gaps, budget signals |\n| `check-in` | Relationship maintenance, satisfaction pulse | NPS proxy, feature requests, referral willingness |\n| `cold-outreach` | Reaching out to potential customers who haven't expressed interest yet | Pain points, willingness to pay signals, competitive mentions |\n\n## Procedure\n\n### Cold Outreach Conversation Type\n\n**Cold Outreach** conversation type:\n- **When**: Reaching out to potential customers who haven't expressed interest yet\n- **Prep**: Research the prospect, prepare problem-led opening, have 3 discovery questions ready\n- **During**: Listen 80%, talk 20%. Goal is learning, not selling.\n- **Debrief**: Capture pain points, willingness to pay signals, competitive mentions\n\n**Starting from Scratch Detection**: If no ICP artifact exists in `artifacts\u002Fproduct\u002F` and no CONV- artifacts exist in `artifacts\u002Fdecision-memos\u002F`, present a \"Starting from scratch\" briefing:\n- Guidance on finding first 10 targets (LinkedIn search, community lurking, competitor reviews)\n- Starter interview script focused on problem discovery\n- Link to icp-positioning skill to formalize learnings after 5+ conversations\n\n### Mom-Test Interview Methodology\n\nWhen conversation type is `mom-test`, apply The Mom Test rules strictly:\n\n**Core Principle**: Talk about their life, not your idea. Extract facts about past behavior, not opinions about the future.\n\n**The 5 Mom-Test Rules**:\n1. **Never ask \"Would you use this?\"** — People are polite liars. Instead ask: \"How do you handle [problem] today?\"\n2. **Never ask \"Would you pay for this?\"** — Instead ask: \"What does [current workaround] cost you in time\u002Fmoney?\"\n3. **Ask about specifics in the past, not hypotheticals** — \"Tell me about the last time [problem] happened\" vs. \"Do you ever experience [problem]?\"\n4. **Talk less, listen more** — Target 80% listening, 20% talking. Follow their energy.\n5. **Extract commitment or advancement** — End every call with a concrete next step (intro to a colleague, LOI review, pilot agreement).\n\n**Mom-Test Question Bank** (use 5-7 per call, 15 min max):\n- \"What's the hardest part of [workflow] for you right now?\"\n- \"Can you walk me through the last time that happened?\"\n- \"What did you do about it? What else have you tried?\"\n- \"How much time\u002Fmoney does that cost you per [week\u002Fmonth]?\"\n- \"Who else in your org deals with this problem?\"\n- \"What would a dream solution look like?\" (only AFTER understanding the problem)\n- \"If you had a magic wand, what would you change about [current tool\u002Fprocess]?\"\n- \"What are you using today to solve this? What's working? What's not?\"\n- \"Have you looked for solutions? What happened?\"\n- \"How urgent is solving this? What's your timeline?\"\n\n**Anti-Patterns to Avoid**:\n- Pitching your solution before minute 10\n- Asking binary yes\u002Fno questions\n- Fishing for compliments (\"Don't you think X would be great?\")\n- Talking about features before understanding the problem\n- Accepting \"that would be nice\" as validation (it's not)\n\n**Debrief Extras for Mom-Test**:\nWhen debriefing a mom-test conversation, also capture:\n- **$$ Quantification**: estimated cost of the problem per month\u002Fyear\n- **Workarounds tried**: what they've already done to solve this\n- **Commitment signal**: did they offer an intro, agree to review an LOI, or ask for a follow-up?\n\n### Mode: `prep \u003Ctype>`\n\n1. **Load Context**\n   - Read `company.config.yaml` for product name, stage, ICP\n   - Check `artifacts\u002Fproduct\u002F` for the most recent ICP positioning artifact\n   - Check `artifacts\u002Fprds\u002F` for in-progress features (gives context on what's being built)\n   - Check recent CONV- artifacts for patterns from this company\u002Fsegment\n\n2. **Generate Pre-Call Briefing**\n\n   Produce a briefing document with:\n\n   **Do Ask** (5 open-ended questions tailored to conversation type):\n   - Discovery: \"What's the hardest part of [workflow] for you right now?\"\n   - Demo: \"Which of these capabilities matters most to your day-to-day?\"\n   - Churn-save: \"What changed since you first signed up?\"\n   - Upsell: \"If you could wave a magic wand and add one capability, what would it be?\"\n   - Check-in: \"What's working well? What's frustrating?\"\n\n   **Don't Ask** (anti-patterns to avoid):\n   - Leading questions (\"Don't you think our feature X is great?\")\n   - Binary questions when you need depth (\"Do you like it?\" → \"What's your experience with it?\")\n   - Feature-first questions before understanding the problem\n\n   **Listen For** (signals specific to conversation type):\n   - ICP fit signals: role, company size, tech stack, budget authority\n   - Competitor mentions: who else they're evaluating or using\n   - Churn risk indicators: frustration level, timeline pressure, alternative mentions\n   - Willingness-to-pay signals: asking about pricing, comparing to current spend\n\n   **Context from Recent Conversations** (if CONV- artifacts exist for this company):\n   - Previous pain points mentioned\n   - Commitments made\n   - Open questions from last call\n\n3. **Present Briefing** — show the briefing to the user before the call. Do not save as an artifact (ephemeral output).\n\n### Mode: `debrief`\n\n4. **Capture Structured Debrief**\n\n   Prompt the user to provide (or extract from conversation notes):\n\n   | Field | Description |\n   |-------|-------------|\n   | **Date** | When the conversation happened |\n   | **Participant** | Name, role, company, segment |\n   | **Conversation type** | discovery \u002F demo \u002F churn-save \u002F upsell \u002F check-in |\n   | **Pain level** | 1 (minor inconvenience) to 5 (hair-on-fire) |\n   | **Key quotes** | 2-3 verbatim quotes that capture the essence |\n   | **Willingness to pay** | Strong signal \u002F Weak signal \u002F No signal \u002F Price sensitive |\n   | **Competitor mentions** | Any alternatives mentioned (by name) |\n   | **Churn signals** | None \u002F Low \u002F Medium \u002F High |\n   | **Upsell signals** | None \u002F Low \u002F Medium \u002F High |\n   | **Feature requests** | Specific capabilities mentioned |\n   | **ICP fit** | Strong \u002F Partial \u002F Weak — with reasoning |\n   | **Recommended next action** | Follow-up call, send proposal, route to support, etc. |\n\n5. **Classify and Tag**\n\n   Based on the debrief, assign tags:\n   - `prd_evidence: true` — if the conversation surfaced a clear product need\n   - `icp_signal: true` — if the conversation provided ICP refinement data\n   - `churn_signal: true` — if churn risk was detected (Medium or High)\n   - `upsell_signal: true` — if expansion opportunity was identified\n\n6. **Save Artifact**\n\n   Write to `artifacts\u002Fdecision-memos\u002FCONV-{date}-{type}-{company}.md`:\n\n   ```yaml\n   ---\n   id: CONV-{date}-{type}-{company}\n   type: decision-memo\n   title: \"{Type} Conversation — {Company}\"\n   status: review\n   created: {date}\n   tags: [customer-conversation, {type}]\n   prd_evidence: true|false\n   icp_signal: true|false\n   churn_signal: true|false\n   ---\n   ```\n\n   Body contains the structured debrief fields above.\n\n7. **Route Signals**\n\n   After saving:\n   - If `churn_signal: true` → inform the user: \"Churn signal detected. Consider running `\u002Fsupport-operations` to review escalation paths, or `\u002Factivation-onboarding` to design a re-engagement flow.\"\n   - If `icp_signal: true` → inform the user: \"ICP data captured. Run `\u002Ficp-positioning` to update your ideal customer profile with this evidence.\"\n   - If `prd_evidence: true` → inform the user: \"Product evidence captured. This will be consumed by `\u002Ffeedback-synthesizer` in the next synthesis cycle.\"\n\n8. **Validate** — run `.\u002Ftools\u002Fartifact\u002Fvalidate.sh` on the saved artifact.\n\n### Mode: `patterns`\n\n9. **Cross-Conversation Analysis**\n\n   Scan all CONV- artifacts in `artifacts\u002Fdecision-memos\u002F`:\n   - Count by conversation type\n   - Extract all `Key quotes` fields\n   - Tally pain levels by segment\n   - List all competitor mentions with frequency\n   - Identify recurring feature requests (3+ mentions = pattern)\n   - Aggregate churn signals by segment\n\n10. **Surface Patterns**\n\n    Present a summary:\n    - **Top pain points**: ranked by frequency × pain level\n    - **Competitor landscape**: who's mentioned most, in what context\n    - **ICP signals**: which segments show strongest fit\n    - **Emerging feature requests**: requests that appear 3+ times\n    - **Churn risk segments**: segments with concentrated churn signals\n\n    This output is ephemeral (presented to user, not saved as artifact). The user can then feed specific findings into `\u002Ffeedback-synthesizer` or `\u002Ficp-positioning`.\n\n## Quality Checklist\n- [ ] Pre-call briefings include 5 open-ended questions (no leading questions)\n- [ ] Debriefs capture all structured fields\n- [ ] Artifact frontmatter uses type `decision-memo` (passes validate.sh)\n- [ ] Signal routing recommendations are provided\n- [ ] Pattern analysis only runs with 3+ CONV- artifacts\n- [ ] Artifact passes validation\n\n## Cross-References\n- **feedback-synthesizer** (S-PROD-04) — consumes CONV- artifacts as input during synthesis\n- **icp-positioning** (S-PROD-01) — CONV- artifacts provide ICP evidence\n- **prd-writer** (S-PROD-02) — CONV- artifacts populate PRD evidence sections\n- **support-operations** (S-OPS-07) — churn signals feed support escalation review\n- **inbound-loop-sop** — `standards\u002Fops\u002Finbound-loop-sop.md` documents the full pipeline\n",{"name":5,"description":7,"argument-hint":30,"user-invocable":31},"[prep \u003Ctype> | debrief | patterns]",true,"main","e842fb6d0e04360c47db78bc2e2e2987314d64d4",".claude\u002Fskills\u002Fcustomer-conversations",2,"https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os","https:\u002F\u002Fgithub.com\u002Fvibbs","Principal Engineer · Applied AI · building ArcAI @ GrowthArc","2026-10-21 02:08:08",{"source":8,"skillName":5,"commands":41,"prompt":51,"zipAvailable":31},[42,45,48],{"tool":43,"command":44},"npx","npx skills add vibbs\u002Fcompany-os --skill customer-conversations",{"tool":46,"command":47},"bunx","bunx skills add vibbs\u002Fcompany-os --skill customer-conversations",{"tool":49,"command":50},"pnpm","pnpm dlx skills add vibbs\u002Fcompany-os --skill customer-conversations","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcustomer-conversations\u002FSKILL.md",[],[54,63,71,81,97,115],{"skillNo":55,"name":56,"nameZh":6,"description":57,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":58,"categories":59,"keywords":60,"publishedAt":25},"01M6CTHBW1E64ECYGD105N4S5E","typo-field","Fixture skill with a misspelled top-level field.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Ftypo-field\u002FSKILL.md",[],[61,62],"typo field","fixture skill",{"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":25},"01M6CTHABTV8FKVDHFMMKPAV51","spaced-comments","Fixture skill with comments and spaces after block keys.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fspaced-comments\u002FSKILL.md",[],[70,62],"spaced comments",{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":25},"01M6CTH8MHC14KB65ZDHYBJW1N","crlf-comments","Fixture skill with Windows line endings.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fcrlf-comments\u002FSKILL.md",[],[78,79,80,62],"crlf comments","CRLF","windows",{"skillNo":82,"name":83,"nameZh":6,"description":84,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":87,"publishedAt":25},"01M6CTH6TBKTDP9G8QB1Z2K0TY","claude-fields","Fixture skill using top-level fields Claude Code documents.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Ftests\u002Ffixtures\u002Fskills\u002Fclaude-fields\u002FSKILL.md",[],[88,89,90,91,92,93,62,94,95,96],"claude fields","explore","documents","src","grep","read","Bash","code","testing",{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":108,"publishedAt":25},"01M6CTH46BGKXH80RWS876Y2CW","workflow-router","Maps an objective to the artifacts, stages and agents it needs. Use when deciding what happens next or which agents to involve.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Fos\u002Fskills\u002Fworkflow-router\u002FSKILL.md",[103],{"categoryNo":104,"categoryName":105,"taxonomyType":106,"taxonomyTypeDesc":107},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[109,110,111,112,113,23,114],"workflow","router","workflow router","next","maps","artifacts",{"skillNo":116,"name":117,"nameZh":6,"description":118,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":119,"categories":120,"keywords":121,"publishedAt":25},"01M6CTH27653SRVP2FMEQN3J5B","weekly-review","Summarizes your week: what shipped, AI spend, support themes, security health and next priorities. Use at the end of each week.","https:\u002F\u002Fgithub.com\u002Fvibbs\u002Fcompany-os\u002Fblob\u002Fmain\u002Fos\u002Fskills\u002Fweekly-review\u002FSKILL.md",[],[122,123,124,112,23,125,126,127,128],"weekly review","review","priorities","health","save","support","security"]