[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6D6118THKJB4QWS3ESSV93A":3,"skill-same:01M6D6118THKJB4QWS3ESSV93A":67},{"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":23,"publishedAt":35,"linkOnly":36,"markdownContent":37,"markdownBody":38,"frontMatter":39,"refName":43,"commitSha":44,"skillDirectory":45,"forksCount":46,"repositoryHtmlUrl":47,"repositoryHomepageUrl":48,"ownerHtmlUrl":49,"ownerBlogUrl":50,"ownerDescription":51,"updatedAt":52,"install":53,"resources":66},"01M6D6118THKJB4QWS3ESSV93A","pmf-engine",null,"Vohra\u002FSupan PMF Engine - evaluate existing user data with Sean Ellis PMF signal, very-disappointed segmentation, retention\u002Fusage evidence, and High-Expectation Customer synthesis","GeorgeQLe\u002Fagentic-skills","GeorgeQLe","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F21134126?v=4","01M6D2JP418S0VGJGF8M93EC3H","George Le","01M6D2JPBAKYD82TA5FS6HKQ3D",1,"MIT","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Fpacks\u002Fbusiness-research\u002Fclaude\u002Fcustomer-discovery\u002Fframeworks\u002Fpmf-engine\u002Farchive\u002Fv0.0\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000021","资料检索与研究","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34],"engine","pmf engine","research","customer","data","segmentation","discovery","user","v0","evidence","Synthesis","2026-11-10 10:12:30",false,"---\nname: pmf-engine\ndescription: Vohra\u002FSupan PMF Engine - evaluate existing user data with Sean Ellis PMF signal, very-disappointed segmentation, retention\u002Fusage evidence, and High-Expectation Customer synthesis\ntype: research\nversion: v0.0\ninvocation: sub-skill\nparent: customer-discovery\n---\n\n## Pack Availability Guard\n\nBefore telling the user to run a skill from another project-local pack, check `.agents\u002Fproject.json.enabled_packs`. If the target pack is not enabled, recommend `\u002Fpack install \u003Cpack>` instead of the target skill. Global skills are always valid. Skills from this same pack are valid because the current skill is already running from that pack.\n\n# PMF Engine - Vohra\u002FSupan Product-Market Fit Analysis\n\n## Report-First Approval Gate\n\nDefault to report-only: present findings, evidence coverage, assumptions, recommended artifact path, and proposed file changes in a pre-approval alignment page plus a concise conversation summary for user approval before creating or updating canonical research, spec, or task files.\n\nDo not write or overwrite synthesized deliverables until the user explicitly approves, unless the user invoked an explicit write\u002Fupdate\u002Ffix mode or clearly asked to write files upfront. Raw evidence capture may be persisted before analysis when reproducibility requires it; report those raw paths separately and still gate synthesized research\u002Freport writes.\n\nWhen stopping for approval, build and attempt to open the alignment preview page first, then ask the user to review it and approve, question, or request adjustments. Do not include `Recommended next skill`, `Recommended next command`, or downstream routing language. The approval request itself is the next action. Only emit next-skill routing after the approved artifact has been written or updated.\n\n## Staged Research Workflow\n\nUse this staged workflow for synthesized research or report outputs that would create or update canonical research, spec, or task files.\n\n1. **Stage 1 - Research and clarify.** Perform the research, run required source\u002Fcode checks, and ask any needed clarification questions. Write only a non-canonical working packet: flat mode uses `research\u002F_working\u002Fpreliminary-pmf-engine-research.md`; product-path mode uses `research\u002F{slug}\u002F_working\u002Fpreliminary-pmf-engine-research.md`. Do not create or update canonical research, spec, or task files in Stage 1.\n2. **Stage 2 - Review alignment.** Consume the working packet and build the `review` HTML alignment page. The page must render the full preliminary packet, evidence matrix, assumptions\u002Fconfidence register, source coverage gaps, proposed canonical file changes, and approval gates. Stop for either feedback-only YAML or final compiled YAML.\n3. **Stage 3 - Finalize approved artifacts.** Consume final compiled YAML only when it has no unresolved negative feedback. Apply approved edits first, archive the working packet, write the approved canonical artifacts, and convert the alignment page to `confirmed`.\n\nGeneric staged path contract: flat mode uses `research\u002F_working\u002Fpreliminary-\u003Cskill>-research.md`; product-path mode uses `research\u002F{slug}\u002F_working\u002Fpreliminary-\u003Cskill>-research.md`. For this skill, replace `\u003Cskill>` with `pmf-engine`. Feedback-only YAML revises the working packet and page, then remains in Stage 2. Finalization requires no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. After approval, archive the working packet to `docs\u002Fhistory\u002Farchive\u002FYYYY-MM-DD\u002FHHMMSS\u002F\u003Coriginal-working-path>`, remove the active working packet, and write approved canonical artifacts only after approval. Canonical output paths remain unchanged.\n\n## Evidence And Feedback Handling\n\nTreat user feedback as input to evaluate, not as automatic ground truth.\n\n- For factual, evidentiary, technical, or source-backed claims: verify against available evidence. Push back clearly and cite the evidence when the user appears to misunderstand.\n- For taste, prioritization, or subjective judgment calls: weigh user feedback heavily and adapt unless it conflicts with verified evidence.\n- When uncertain, say what is known, what is inferred, and what would change the conclusion.\n\n## Prerequisites\n\n- **Hard**: ICP candidates from orchestrator working packet (`research\u002F_working\u002Fpreliminary-customer-discovery-research.md`) or `research\u002Ficp.md` (or `research\u002F{slug}\u002Ficp.md`) must exist. If neither exists, tell the user to run `\u002Fcustomer-discovery` first and stop.\n- **Hard**: Real user or customer evidence must exist. Acceptable evidence includes `research\u002Fcustomer-feedback.md`, survey\u002Finterview data, support tickets, churn\u002Fretention notes, product analytics, sales calls, usage logs, or a user-provided dataset. If no real-user evidence exists, explain that PMF Engine is unavailable for pre-product work and stop.\n- **Soft**: `research\u002Fcompetitive-analysis.md`, specs, product analytics exports, NPS\u002FCSAT reports, cohort retention data, customer support records, sales notes, and codebase context when available.\n\n## Process\n\n### 0. Product-Path Scope Resolution\n\nResolve research scope by product path before using code or app structure as a hint:\n\n1. If `$ARGUMENTS` names a non-archived `research\u002F{slug}\u002F` directory or a product-path ID whose `scope_path` points there, use that path. Treat `{slug}` as the product\u002Fapp name, not the ICP, audience, or segment label.\n2. If `$ARGUMENTS` names only `research\u002F_archive\u002F{slug}\u002F` or a manifest entry with `status: archived` or legacy `status: abandoned`, stop and warn that the path is archived; do not write or update scoped outputs there.\n3. Read `research\u002F.progress.yaml` when present. Normalize legacy `active_path` to `active_paths` on read and write back `active_paths` on manifest updates. Treat legacy `abandoned` as `archived`; exclude `archived`, `abandoned`, `deferred`, `revisit_candidate`, `promoted`, and any `scope_path` under `research\u002F_archive\u002F` from active target selection.\n4. If active product paths exist in the manifest, use those paths. If multiple active paths exist, ask which one to target unless this skill explicitly supports cross-path output.\n5. If no active manifest target exists, list non-archived product directories under `research\u002F`, excluding `research\u002F_archive\u002F` and dot directories. Auto-select only when exactly one exists; ask when multiple exist.\n6. If no product directories exist, use flat `research\u002F` single-product mode.\n7. Detect monorepo\u002Fapp\u002Fpackage structure only as a secondary hint.\n\nWhen product path `{slug}` is active, read and write research under `research\u002F{slug}\u002F`, specs under `specs\u002F{slug}\u002F`.\n\n### 1. Load Context And Evidence\n\n- Read ICP candidates from orchestrator working packet (`research\u002F_working\u002Fpreliminary-customer-discovery-research.md`) or `research\u002Ficp.md`.\n- Read real-user evidence: `research\u002Fcustomer-feedback.md`, survey results, interview notes, support tickets, sales notes, churn notes, retention\u002Fcohort reports, product analytics, usage logs, or user-provided datasets.\n- Read idea brief, competitive analysis, specs, README, CLAUDE.md, and codebase context when available.\n- Classify evidence by source type, recency, sample size, segment coverage, and bias risk.\n- Separate actual observed user\u002Fcustomer data from founder hypotheses, analyst inference, and marketing claims.\n\n### 2. Data Readiness And PMF Survey Coverage\n\nAssess whether the evidence can support the Vohra\u002FSuperhuman PMF Engine and HXC synthesis:\n\n**PMF survey availability**\n- Whether the Sean Ellis question is present: \"How would you feel if you could no longer use this product?\"\n- Response distribution across Very disappointed, Somewhat disappointed, Not disappointed, and N\u002FA.\n- Sample size, response bias, segment coverage, and whether respondents are active users.\n- Whether supporting questions exist: who would benefit most, main benefit received, alternatives used, and desired improvements.\n\n**Behavioral evidence availability**\n- Retention, activation, frequency, depth of use, expansion, repeat purchase, or renewal data.\n- Qualitative evidence from interviews, support, churn reasons, sales calls, and onboarding notes.\n- Segment-level differences between high-love users and lukewarm or churned users.\n\n**Data-readiness verdict**\n- Classify as Strong \u002F Partial \u002F Insufficient.\n- If insufficient, identify the minimum evidence needed before PMF Engine can produce a reliable verdict.\n\n### 3. Sean Ellis PMF Signal\n\nWhen survey data exists, calculate and interpret PMF signal:\n\n**Very-disappointed threshold**\n- Calculate the percentage of respondents who would be Very disappointed.\n- Treat 40%+ as a positive PMF signal, but do not treat it as sufficient by itself.\n- Segment the signal by ICP candidate, role, company type, use case, acquisition source, tenure, and engagement where evidence allows.\n\n**Signal quality**\n- Check whether respondents are representative of the target user base.\n- Identify inflated signal risks: tiny sample size, only enthusiasts responding, founder-picked respondents, inactive-user exclusion, or ambiguous product definition.\n- Identify hidden weak spots: users love one workflow but reject the broader product, decision-makers differ from daily users, or retention contradicts survey love.\n\n### 4. Very-Disappointed Segment Analysis\n\nAnalyze the \"very disappointed\" cohort to identify who gets irreplaceable value:\n\n**Segment attributes**\n- Role, company type, size, maturity, workflow, urgency, technical readiness, buying context, and product use case.\n- Acquisition path and activation path.\n- Retention, expansion, support burden, and engagement patterns.\n\n**Loved benefit**\n- The specific outcome, workflow, or emotional relief that makes the product hard to lose.\n- User language for the benefit, not only agent summaries.\n- Differences between what users say they value and what their behavior indicates.\n\n**Alternatives and switching**\n- What users would do without the product.\n- Which alternatives are inadequate and why.\n- Whether users are switching from nothing, manual workarounds, an incumbent tool, or another team member's labor.\n\n### 5. High-Expectation Customer Synthesis\n\nSynthesize the High-Expectation Customer (HXC): the most discerning customer within the target market who still gets exceptional value.\n\n**HXC profile**\n- Who they are, what they are trying to accomplish, and why their expectations are high.\n- What makes them hard to satisfy and why the product still works for them.\n- What they refuse to tolerate: workflow friction, missing integrations, quality gaps, risk, pricing friction, or trust issues.\n\n**HXC value wedge**\n- The narrow product promise that this customer would miss most.\n- Why the promise is meaningfully better than alternatives for this segment.\n- Which product qualities must remain non-negotiable to keep the HXC.\n\n**Exclusion logic**\n- Adjacent users or buyers who look attractive but do not fit the HXC.\n- Segments with high willingness but poor retention, high usage but no budget, or strong pain but no realistic success path.\n\n### 6. Target Segment And Product Focus\n\nUse PMF signal and HXC synthesis to narrow ICP recommendations:\n\n**Primary segment**\n- Recommend the segment with the strongest combination of PMF signal, behavioral evidence, HXC fit, and business viability.\n- Explain whether the segment is ready for positioning, requires more validation, or should remain provisional.\n\n**Product focus**\n- Identify the features, workflows, or promises to double down on for the target segment.\n- Identify requests to ignore, defer, or treat as segment-specific because they dilute the loved benefit.\n- Identify onboarding, activation, or messaging changes that could increase very-disappointed share.\n\n**Risks**\n- Retention contradiction, buyer\u002Fuser mismatch, sample-size weakness, insufficient product maturity, low willingness to pay, or overfitting to a small enthusiastic group.\n\n### 7. Validate With User\n\nPresent PMF signal, very-disappointed segment analysis, HXC synthesis, and target-segment recommendation. Ask for corrections on:\n- Evidence completeness: missing surveys, interviews, analytics, churn notes, or support records.\n- Segment accuracy: whether the high-love segment matches known users and buyers.\n- HXC realism: whether the proposed HXC is demanding enough and still reachable.\n- Product focus: whether the recommended double-downs and exclusions match known product strategy.\n\n## Output\n\n### `research\u002Fcustomer-discovery-pmf-engine.md` (or `research\u002F{slug}\u002Fcustomer-discovery-pmf-engine.md`)\n\n```markdown\n# PMF Engine Analysis\n\n> Based on: [sources]\n> Date: [current date]\n> Methodology: Vohra\u002FSupan PMF Engine (Sean Ellis PMF signal, Very-Disappointed segmentation, High-Expectation Customer synthesis)\n\n## Data Readiness\n| Evidence Type | Available? | Sample \u002F Coverage | Bias Risk | Notes |\n|---------------|------------|-------------------|-----------|-------|\n| PMF survey | Yes\u002FNo\u002FPartial | [count\u002Fsegments] | Low\u002FMedium\u002FHigh | [notes] |\n| Behavioral usage | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n| Retention\u002Fchurn | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n| Customer interviews\u002Fsupport | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n\n## PMF Signal\n| Segment | Very Disappointed % | Sample Size | Supportive Behavioral Evidence | Confidence |\n|---------|---------------------|-------------|--------------------------------|------------|\n| [segment] | [percent] | [n] | [retention\u002Fusage\u002Frenewal] | High\u002FMedium\u002FLow |\n\n## Very-Disappointed Segment\n### Segment Attributes\n- [role\u002Fcompany\u002Fuse case\u002Fmaturity\u002Facquisition path]\n\n### Loved Benefit\n- [user-language description of the benefit users would miss]\n\n### Alternatives\n| Alternative | Why Users Would Switch There | Why It Falls Short | Evidence |\n|-------------|------------------------------|--------------------|----------|\n| [alternative] | [reason] | [gap] | [source] |\n\n## High-Expectation Customer\n### HXC Profile\n- **Customer**: [specific profile]\n- **High expectations**: [what makes them demanding]\n- **Why this product works**: [evidence-backed reason]\n- **Non-negotiables**: [quality\u002Fworkflow\u002Ftrust requirements]\n\n### HXC Value Wedge\n- [narrow promise and why it is better than alternatives]\n\n### Exclusions\n| Excluded Segment | Why It Looks Attractive | Why It Is Not HXC | Evidence |\n|------------------|-------------------------|-------------------|----------|\n| [segment] | [signal] | [fit issue] | [source] |\n\n## Target Segment Recommendation\n- **Primary target**: [segment]\n- **Verdict**: Strong PMF signal \u002F partial signal \u002F insufficient evidence\n- **Reasoning**: [evidence-backed rationale]\n- **Product focus**: [double-downs]\n- **Defer or ignore**: [requests\u002Fsegments\u002Ffeatures that dilute focus]\n\n## Evidence Matrix\n| Claim | Evidence Source | Evidence Type | Confidence |\n|-------|-----------------|---------------|------------|\n| [claim] | [source] | Observed\u002FInferred\u002FHypothesized | High\u002FMedium\u002FLow |\n\n## Open Evidence Gaps\n- [gap and what would change the recommendation]\n```\n\n## Constraints\n\n- Use PMF Engine only for product-exists contexts with real user\u002Fcustomer evidence. Do not fabricate PMF signal from market research or founder hypotheses.\n- Do not treat the 40% very-disappointed threshold as a standalone verdict; cross-check it against sample quality, retention, engagement, and segment viability.\n- Preserve the distinction between customer, buyer, user, and HXC.\n- Ground every segment recommendation, exclusion, and product-focus recommendation in evidence.\n- Do not prescribe UI or architecture - describe PMF, segment focus, and customer evidence implications.\n- Present findings before writing.\n- Do not overwrite existing output without asking the user first.\n- This is a sub-skill - do not emit next-step routing.\n\n## Alignment Page\n\nWhen this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following `ALIGNMENT-PAGE.md` in this skill's directory. Output: `alignment\u002Fpmf-engine-{topic}.html`.\n\n## Default Shipping Contract\n\nFollow the shared shipping contract convention in CLAUDE.md.\n","## Pack Availability Guard\n\nBefore telling the user to run a skill from another project-local pack, check `.agents\u002Fproject.json.enabled_packs`. If the target pack is not enabled, recommend `\u002Fpack install \u003Cpack>` instead of the target skill. Global skills are always valid. Skills from this same pack are valid because the current skill is already running from that pack.\n\n# PMF Engine - Vohra\u002FSupan Product-Market Fit Analysis\n\n## Report-First Approval Gate\n\nDefault to report-only: present findings, evidence coverage, assumptions, recommended artifact path, and proposed file changes in a pre-approval alignment page plus a concise conversation summary for user approval before creating or updating canonical research, spec, or task files.\n\nDo not write or overwrite synthesized deliverables until the user explicitly approves, unless the user invoked an explicit write\u002Fupdate\u002Ffix mode or clearly asked to write files upfront. Raw evidence capture may be persisted before analysis when reproducibility requires it; report those raw paths separately and still gate synthesized research\u002Freport writes.\n\nWhen stopping for approval, build and attempt to open the alignment preview page first, then ask the user to review it and approve, question, or request adjustments. Do not include `Recommended next skill`, `Recommended next command`, or downstream routing language. The approval request itself is the next action. Only emit next-skill routing after the approved artifact has been written or updated.\n\n## Staged Research Workflow\n\nUse this staged workflow for synthesized research or report outputs that would create or update canonical research, spec, or task files.\n\n1. **Stage 1 - Research and clarify.** Perform the research, run required source\u002Fcode checks, and ask any needed clarification questions. Write only a non-canonical working packet: flat mode uses `research\u002F_working\u002Fpreliminary-pmf-engine-research.md`; product-path mode uses `research\u002F{slug}\u002F_working\u002Fpreliminary-pmf-engine-research.md`. Do not create or update canonical research, spec, or task files in Stage 1.\n2. **Stage 2 - Review alignment.** Consume the working packet and build the `review` HTML alignment page. The page must render the full preliminary packet, evidence matrix, assumptions\u002Fconfidence register, source coverage gaps, proposed canonical file changes, and approval gates. Stop for either feedback-only YAML or final compiled YAML.\n3. **Stage 3 - Finalize approved artifacts.** Consume final compiled YAML only when it has no unresolved negative feedback. Apply approved edits first, archive the working packet, write the approved canonical artifacts, and convert the alignment page to `confirmed`.\n\nGeneric staged path contract: flat mode uses `research\u002F_working\u002Fpreliminary-\u003Cskill>-research.md`; product-path mode uses `research\u002F{slug}\u002F_working\u002Fpreliminary-\u003Cskill>-research.md`. For this skill, replace `\u003Cskill>` with `pmf-engine`. Feedback-only YAML revises the working packet and page, then remains in Stage 2. Finalization requires no unresolved `needs-clarification`, unresolved `down` feedback, or other unresolved negative feedback. After approval, archive the working packet to `docs\u002Fhistory\u002Farchive\u002FYYYY-MM-DD\u002FHHMMSS\u002F\u003Coriginal-working-path>`, remove the active working packet, and write approved canonical artifacts only after approval. Canonical output paths remain unchanged.\n\n## Evidence And Feedback Handling\n\nTreat user feedback as input to evaluate, not as automatic ground truth.\n\n- For factual, evidentiary, technical, or source-backed claims: verify against available evidence. Push back clearly and cite the evidence when the user appears to misunderstand.\n- For taste, prioritization, or subjective judgment calls: weigh user feedback heavily and adapt unless it conflicts with verified evidence.\n- When uncertain, say what is known, what is inferred, and what would change the conclusion.\n\n## Prerequisites\n\n- **Hard**: ICP candidates from orchestrator working packet (`research\u002F_working\u002Fpreliminary-customer-discovery-research.md`) or `research\u002Ficp.md` (or `research\u002F{slug}\u002Ficp.md`) must exist. If neither exists, tell the user to run `\u002Fcustomer-discovery` first and stop.\n- **Hard**: Real user or customer evidence must exist. Acceptable evidence includes `research\u002Fcustomer-feedback.md`, survey\u002Finterview data, support tickets, churn\u002Fretention notes, product analytics, sales calls, usage logs, or a user-provided dataset. If no real-user evidence exists, explain that PMF Engine is unavailable for pre-product work and stop.\n- **Soft**: `research\u002Fcompetitive-analysis.md`, specs, product analytics exports, NPS\u002FCSAT reports, cohort retention data, customer support records, sales notes, and codebase context when available.\n\n## Process\n\n### 0. Product-Path Scope Resolution\n\nResolve research scope by product path before using code or app structure as a hint:\n\n1. If `$ARGUMENTS` names a non-archived `research\u002F{slug}\u002F` directory or a product-path ID whose `scope_path` points there, use that path. Treat `{slug}` as the product\u002Fapp name, not the ICP, audience, or segment label.\n2. If `$ARGUMENTS` names only `research\u002F_archive\u002F{slug}\u002F` or a manifest entry with `status: archived` or legacy `status: abandoned`, stop and warn that the path is archived; do not write or update scoped outputs there.\n3. Read `research\u002F.progress.yaml` when present. Normalize legacy `active_path` to `active_paths` on read and write back `active_paths` on manifest updates. Treat legacy `abandoned` as `archived`; exclude `archived`, `abandoned`, `deferred`, `revisit_candidate`, `promoted`, and any `scope_path` under `research\u002F_archive\u002F` from active target selection.\n4. If active product paths exist in the manifest, use those paths. If multiple active paths exist, ask which one to target unless this skill explicitly supports cross-path output.\n5. If no active manifest target exists, list non-archived product directories under `research\u002F`, excluding `research\u002F_archive\u002F` and dot directories. Auto-select only when exactly one exists; ask when multiple exist.\n6. If no product directories exist, use flat `research\u002F` single-product mode.\n7. Detect monorepo\u002Fapp\u002Fpackage structure only as a secondary hint.\n\nWhen product path `{slug}` is active, read and write research under `research\u002F{slug}\u002F`, specs under `specs\u002F{slug}\u002F`.\n\n### 1. Load Context And Evidence\n\n- Read ICP candidates from orchestrator working packet (`research\u002F_working\u002Fpreliminary-customer-discovery-research.md`) or `research\u002Ficp.md`.\n- Read real-user evidence: `research\u002Fcustomer-feedback.md`, survey results, interview notes, support tickets, sales notes, churn notes, retention\u002Fcohort reports, product analytics, usage logs, or user-provided datasets.\n- Read idea brief, competitive analysis, specs, README, CLAUDE.md, and codebase context when available.\n- Classify evidence by source type, recency, sample size, segment coverage, and bias risk.\n- Separate actual observed user\u002Fcustomer data from founder hypotheses, analyst inference, and marketing claims.\n\n### 2. Data Readiness And PMF Survey Coverage\n\nAssess whether the evidence can support the Vohra\u002FSuperhuman PMF Engine and HXC synthesis:\n\n**PMF survey availability**\n- Whether the Sean Ellis question is present: \"How would you feel if you could no longer use this product?\"\n- Response distribution across Very disappointed, Somewhat disappointed, Not disappointed, and N\u002FA.\n- Sample size, response bias, segment coverage, and whether respondents are active users.\n- Whether supporting questions exist: who would benefit most, main benefit received, alternatives used, and desired improvements.\n\n**Behavioral evidence availability**\n- Retention, activation, frequency, depth of use, expansion, repeat purchase, or renewal data.\n- Qualitative evidence from interviews, support, churn reasons, sales calls, and onboarding notes.\n- Segment-level differences between high-love users and lukewarm or churned users.\n\n**Data-readiness verdict**\n- Classify as Strong \u002F Partial \u002F Insufficient.\n- If insufficient, identify the minimum evidence needed before PMF Engine can produce a reliable verdict.\n\n### 3. Sean Ellis PMF Signal\n\nWhen survey data exists, calculate and interpret PMF signal:\n\n**Very-disappointed threshold**\n- Calculate the percentage of respondents who would be Very disappointed.\n- Treat 40%+ as a positive PMF signal, but do not treat it as sufficient by itself.\n- Segment the signal by ICP candidate, role, company type, use case, acquisition source, tenure, and engagement where evidence allows.\n\n**Signal quality**\n- Check whether respondents are representative of the target user base.\n- Identify inflated signal risks: tiny sample size, only enthusiasts responding, founder-picked respondents, inactive-user exclusion, or ambiguous product definition.\n- Identify hidden weak spots: users love one workflow but reject the broader product, decision-makers differ from daily users, or retention contradicts survey love.\n\n### 4. Very-Disappointed Segment Analysis\n\nAnalyze the \"very disappointed\" cohort to identify who gets irreplaceable value:\n\n**Segment attributes**\n- Role, company type, size, maturity, workflow, urgency, technical readiness, buying context, and product use case.\n- Acquisition path and activation path.\n- Retention, expansion, support burden, and engagement patterns.\n\n**Loved benefit**\n- The specific outcome, workflow, or emotional relief that makes the product hard to lose.\n- User language for the benefit, not only agent summaries.\n- Differences between what users say they value and what their behavior indicates.\n\n**Alternatives and switching**\n- What users would do without the product.\n- Which alternatives are inadequate and why.\n- Whether users are switching from nothing, manual workarounds, an incumbent tool, or another team member's labor.\n\n### 5. High-Expectation Customer Synthesis\n\nSynthesize the High-Expectation Customer (HXC): the most discerning customer within the target market who still gets exceptional value.\n\n**HXC profile**\n- Who they are, what they are trying to accomplish, and why their expectations are high.\n- What makes them hard to satisfy and why the product still works for them.\n- What they refuse to tolerate: workflow friction, missing integrations, quality gaps, risk, pricing friction, or trust issues.\n\n**HXC value wedge**\n- The narrow product promise that this customer would miss most.\n- Why the promise is meaningfully better than alternatives for this segment.\n- Which product qualities must remain non-negotiable to keep the HXC.\n\n**Exclusion logic**\n- Adjacent users or buyers who look attractive but do not fit the HXC.\n- Segments with high willingness but poor retention, high usage but no budget, or strong pain but no realistic success path.\n\n### 6. Target Segment And Product Focus\n\nUse PMF signal and HXC synthesis to narrow ICP recommendations:\n\n**Primary segment**\n- Recommend the segment with the strongest combination of PMF signal, behavioral evidence, HXC fit, and business viability.\n- Explain whether the segment is ready for positioning, requires more validation, or should remain provisional.\n\n**Product focus**\n- Identify the features, workflows, or promises to double down on for the target segment.\n- Identify requests to ignore, defer, or treat as segment-specific because they dilute the loved benefit.\n- Identify onboarding, activation, or messaging changes that could increase very-disappointed share.\n\n**Risks**\n- Retention contradiction, buyer\u002Fuser mismatch, sample-size weakness, insufficient product maturity, low willingness to pay, or overfitting to a small enthusiastic group.\n\n### 7. Validate With User\n\nPresent PMF signal, very-disappointed segment analysis, HXC synthesis, and target-segment recommendation. Ask for corrections on:\n- Evidence completeness: missing surveys, interviews, analytics, churn notes, or support records.\n- Segment accuracy: whether the high-love segment matches known users and buyers.\n- HXC realism: whether the proposed HXC is demanding enough and still reachable.\n- Product focus: whether the recommended double-downs and exclusions match known product strategy.\n\n## Output\n\n### `research\u002Fcustomer-discovery-pmf-engine.md` (or `research\u002F{slug}\u002Fcustomer-discovery-pmf-engine.md`)\n\n```markdown\n# PMF Engine Analysis\n\n> Based on: [sources]\n> Date: [current date]\n> Methodology: Vohra\u002FSupan PMF Engine (Sean Ellis PMF signal, Very-Disappointed segmentation, High-Expectation Customer synthesis)\n\n## Data Readiness\n| Evidence Type | Available? | Sample \u002F Coverage | Bias Risk | Notes |\n|---------------|------------|-------------------|-----------|-------|\n| PMF survey | Yes\u002FNo\u002FPartial | [count\u002Fsegments] | Low\u002FMedium\u002FHigh | [notes] |\n| Behavioral usage | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n| Retention\u002Fchurn | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n| Customer interviews\u002Fsupport | Yes\u002FNo\u002FPartial | [coverage] | Low\u002FMedium\u002FHigh | [notes] |\n\n## PMF Signal\n| Segment | Very Disappointed % | Sample Size | Supportive Behavioral Evidence | Confidence |\n|---------|---------------------|-------------|--------------------------------|------------|\n| [segment] | [percent] | [n] | [retention\u002Fusage\u002Frenewal] | High\u002FMedium\u002FLow |\n\n## Very-Disappointed Segment\n### Segment Attributes\n- [role\u002Fcompany\u002Fuse case\u002Fmaturity\u002Facquisition path]\n\n### Loved Benefit\n- [user-language description of the benefit users would miss]\n\n### Alternatives\n| Alternative | Why Users Would Switch There | Why It Falls Short | Evidence |\n|-------------|------------------------------|--------------------|----------|\n| [alternative] | [reason] | [gap] | [source] |\n\n## High-Expectation Customer\n### HXC Profile\n- **Customer**: [specific profile]\n- **High expectations**: [what makes them demanding]\n- **Why this product works**: [evidence-backed reason]\n- **Non-negotiables**: [quality\u002Fworkflow\u002Ftrust requirements]\n\n### HXC Value Wedge\n- [narrow promise and why it is better than alternatives]\n\n### Exclusions\n| Excluded Segment | Why It Looks Attractive | Why It Is Not HXC | Evidence |\n|------------------|-------------------------|-------------------|----------|\n| [segment] | [signal] | [fit issue] | [source] |\n\n## Target Segment Recommendation\n- **Primary target**: [segment]\n- **Verdict**: Strong PMF signal \u002F partial signal \u002F insufficient evidence\n- **Reasoning**: [evidence-backed rationale]\n- **Product focus**: [double-downs]\n- **Defer or ignore**: [requests\u002Fsegments\u002Ffeatures that dilute focus]\n\n## Evidence Matrix\n| Claim | Evidence Source | Evidence Type | Confidence |\n|-------|-----------------|---------------|------------|\n| [claim] | [source] | Observed\u002FInferred\u002FHypothesized | High\u002FMedium\u002FLow |\n\n## Open Evidence Gaps\n- [gap and what would change the recommendation]\n```\n\n## Constraints\n\n- Use PMF Engine only for product-exists contexts with real user\u002Fcustomer evidence. Do not fabricate PMF signal from market research or founder hypotheses.\n- Do not treat the 40% very-disappointed threshold as a standalone verdict; cross-check it against sample quality, retention, engagement, and segment viability.\n- Preserve the distinction between customer, buyer, user, and HXC.\n- Ground every segment recommendation, exclusion, and product-focus recommendation in evidence.\n- Do not prescribe UI or architecture - describe PMF, segment focus, and customer evidence implications.\n- Present findings before writing.\n- Do not overwrite existing output without asking the user first.\n- This is a sub-skill - do not emit next-step routing.\n\n## Alignment Page\n\nWhen this skill produces durable deliverables (research, specs, plans, reports, prototypes, or any document output), build a full-depth HTML alignment page following `ALIGNMENT-PAGE.md` in this skill's directory. Output: `alignment\u002Fpmf-engine-{topic}.html`.\n\n## Default Shipping Contract\n\nFollow the shared shipping contract convention in CLAUDE.md.\n",{"name":5,"type":26,"parent":40,"version":41,"invocation":42,"description":7},"customer-discovery","v0.0","sub-skill","master","96e2912af886dd1606b1b97df707715aef062b02","packs\u002Fbusiness-research\u002Fclaude\u002Fcustomer-discovery\u002Fframeworks\u002Fpmf-engine\u002Farchive\u002Fv0.0",0,"https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills","https:\u002F\u002Fagentic-skills-blush.vercel.app","https:\u002F\u002Fgithub.com\u002FGeorgeQLe","georgeqle.com","George Le is a futurist and serial tech entrepreneur who wants to work toward a world where humanity is free from robotic tasks.","2026-11-10 04:07:24",{"source":8,"skillName":5,"commands":54,"prompt":64,"zipAvailable":65},[55,58,61],{"tool":56,"command":57},"npx","npx skills add GeorgeQLe\u002Fagentic-skills --skill pmf-engine",{"tool":59,"command":60},"bunx","bunx skills add GeorgeQLe\u002Fagentic-skills --skill pmf-engine",{"tool":62,"command":63},"pnpm","pnpm dlx skills add GeorgeQLe\u002Fagentic-skills --skill pmf-engine","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Fpacks\u002Fbusiness-research\u002Fclaude\u002Fcustomer-discovery\u002Fframeworks\u002Fpmf-engine\u002Farchive\u002Fv0.0\u002FSKILL.md",true,[],[68,80,90,102,115,126],{"skillNo":69,"name":70,"nameZh":6,"description":71,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":35},"01M6G0NJES15RJKZE6YZ0TSYGF","source-checkout-allowed","Valid source-checkout-only fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fsource-checkout-allowed\u002FSKILL.md",[],[75,76,77,78,32,79],"source checkout allowed","checkout","fixture","only","valid",{"skillNo":81,"name":82,"nameZh":6,"description":83,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":35},"01M6G0NFZ3R1MNB1E3DVTM0WB7","deck-install-valid","Valid deck install fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fdeck-install\u002FSKILL.md",[],[87,88,89,79,77,32],"deck install valid","deck","install",{"skillNo":91,"name":92,"nameZh":6,"description":93,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":96,"publishedAt":35},"01M6G0ND3QZQC4MDZ9NVB97PT2","codex-pack-guard-valid","Valid Codex Pack Availability Guard fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-pack-guard\u002FSKILL.md",[],[97,98,99,79,100,101,77,32],"codex pack guard valid","pack","guard","codex","availability",{"skillNo":103,"name":104,"nameZh":6,"description":105,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":106,"categories":107,"keywords":111,"publishedAt":35},"01M6G0NA9YB3NKGZKD3S96JMNB","codex-missing-skill-valid","Valid Codex missing skill fallback fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-missing-skill-fallback\u002FSKILL.md",[108],{"categoryNo":109,"categoryName":110,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[112,113,79,100,32,114,77],"codex missing skill valid","missing","fallback",{"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":122,"publishedAt":35},"01M6G0N7PC0EKFD40ZWQAXSAQ9","codex-base-skill-fallback-valid","Valid Codex base skill fallback fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fcodex-base-skill-fallback\u002FSKILL.md",[121],{"categoryNo":109,"categoryName":110,"taxonomyType":21,"taxonomyTypeDesc":22},[79,100,114,123,124,125,32,77],"base skill","codex base skill fallback valid","base",{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":35},"01M6G0N4VRR2YKZQ6N3Y6S2BQA","claude-pack-guard-valid","Valid Claude Pack Availability Guard fixture.","https:\u002F\u002Fgithub.com\u002FGeorgeQLe\u002Fagentic-skills\u002Fblob\u002Fmaster\u002Ftests\u002Ffixtures\u002Fskill-install-routing\u002Fvalid\u002Fclaude-pack-guard\u002FSKILL.md",[],[79,99,98,133,101,32,77],"claude pack guard valid"]