create-dss-project 0.1.0

package/template/skills/onboard/SKILL.md

@@ -0,0 +1,434 @@
+---
+name: onboard
+description: One-time setup skill. Guides the user through a series of questions to configure this project template for their specific strategy project. Run once at the start of a new project, then delete or archive this skill.
+---
+
+# Project Onboarding
+
+**One-time use** — run this when starting a new project from the strategy template. It replaces all `{{PLACEHOLDER}}` values and configures optional modules.
+
+---
+
+## Setup Mode Selection
+
+Before starting, ask the user which mode they prefer:
+
+> **Quick Start** or **Full Setup**?
+>
+> - **Quick Start** — 7 questions only. I'll auto-fill sensible defaults for everything else. Good if you want to get going fast and fine-tune later.
+> - **Full Setup** — All questions across all groups. Full control over every setting.
+
+### Quick Start Mode
+
+Ask only these 7 questions:
+
+1. **What's the project name?** (e.g., "Immutable Audience — Mobile")
+2. **One-line description** — what is this project about in one sentence? (e.g., "Evaluate the mobile gaming market for audience monetisation opportunities")
+3. **What's the core hypothesis or strategic bet?** (e.g., "Mobile gamers aged 18–30 will pay for premium audience insights if delivered in-app")
+4. **What type of project is this?**
+   - **Market entry** — evaluating a new market or product opportunity
+   - **Growth strategy** — expanding an existing product into new channels or segments
+   - **Competitor research** — mapping the competitive landscape and intelligence gathering
+   - **Product launch / GTM** — bringing a product or feature to market
+   - **Internal implementation** — rolling out a new system, process, or initiative
+   - **Vendor / partner evaluation** — selecting tools, platforms, or partners
+   - **Due diligence** — M&A, investment, or acquisition evaluation
+   - **Business case** — building a case for investment or strategic change
+   - **Transformation / change** — organisational or process transformation
+   - **Custom** — something else (you'll choose which modules to include)
+5. **What's the primary entity you're tracking?** (e.g., "studio", "customer", "partner") — enter "none" to skip the discovery module entirely
+
+6. **What's your main goal?** (Pick one)
+   - **Learning** — I want to explore the template and understand how it works
+   - **Real project** — I have a specific strategy project to run
+   - **Team coordination** — I need a shared workspace for my team's strategy work
+   - **Exploration** — I'm evaluating whether this tool fits my needs
+
+7. **How much structure do you want?**
+   - **Full** — All features: scoring, kill conditions, evidence grading, weekly reports, context snapshots
+   - **Essentials** — Core features: research, pipeline, decisions, dashboard (skip scoring, kill conditions, evidence grading)
+   - **Minimal** — Bare bones: research and notes only (skip pipeline, scoring, kill conditions, dashboard)
+
+#### Goal-based guidance
+
+When goal is **Learning**:
+- Suggest browsing the getting-started guide: "See `docs/getting-started.md` for a walkthrough."
+
+#### Structure-based configuration
+
+When structure is **Full**:
+- Proceed with all defaults as normal.
+
+When structure is **Essentials**:
+- Skip Group 5 (scoring dimensions) and Group 6 (kill conditions)
+- Set default scoring dimensions but don't ask
+- Set kill conditions to empty (can be added later)
+- Remove `/compare-options` from CLAUDE.md skills table reference
+- Set features in `project.config.json`: `scoring: false`, `killConditions: false`, `evidenceGrading: false`, `weeklyReports: true`, `contextSnapshots: true`
+
+When structure is **Minimal**:
+- Skip Groups 3–6 entirely
+- Disable discovery module, pipeline, dashboard, scoring, kill conditions
+- Only keep: `research/`, `memory/MEMORY.md`, `memory/research.md`, `memory/decisions.md`, `docs/`
+- Set features in `project.config.json`: `scoring: false`, `killConditions: false`, `evidenceGrading: false`, `weeklyReports: false`, `contextSnapshots: false`
+
+After auto-filling defaults for **Essentials** or **Minimal**, tell the user: "You can always add more features later by editing `project.config.json` and running `/health-check`."
+
+Then auto-fill defaults:
+- `{{GOAL}}` → "Validate hypothesis and produce a go/no-go recommendation within 8 weeks"
+- `{{TEAM}}` → "TBD — update in MEMORY.md"
+- `{{SCOPE_DEFINITION}}` → "Defined by hypothesis — refine after initial discovery"
+- `{{OUT_OF_SCOPE}}` → "TBD — update in MEMORY.md"
+- `{{ICP_DESCRIPTION}}` → "TBD — update in MEMORY.md"
+- `{{PIPELINE_SOURCE_OF_TRUTH}}` → `data/entities.csv`
+- `{{DASHBOARD_URL}}` → "TBD"
+- Scoring dimensions → based on project type (see Group 5 defaults)
+- Kill conditions → "TBD — define within first week"
+- Discovery module → included unless user answered "none" for entity type (or "Minimal" structure)
+- Dashboard module → included by default (unless "Minimal" structure)
+
+After auto-filling, proceed directly to the **Validation / Dry-Run Step**, then continue with replacement.
+
+### Full Setup Mode
+
+Run the complete question flow below (Group 0 through Group 6, including Groups 1A and 3A).
+
+---
+
+## How to Run
+
+Tell the user: "I'll walk you through setting up this project. I'll ask a series of questions, then update all template files with your answers."
+
+---
+
+## Question Flow
+
+Ask these questions **one group at a time**. Wait for answers before proceeding.
+
+### Group 0: Prerequisites Check
+
+Before asking any project questions, verify the environment is ready.
+
+1. **Check Node.js is installed**:
+   ```bash
+   node --version
+   ```
+   If not found, warn: "Node.js is required for building the dashboard. Install it from https://nodejs.org before continuing, or skip the dashboard module."
+
+2. **Check for fresh template**:
+   Scan `memory/MEMORY.md` for any non-placeholder content — if the template placeholders (`{{PROJECT_NAME}}`, `{{GOAL}}`, etc.) are already gone (i.e., replaced with real values), warn:
+   > "This doesn't look like a fresh template — MEMORY.md already contains non-placeholder content. Are you sure you want to continue? This may overwrite existing project data."
+
+   Wait for confirmation before proceeding.
+
+3. **Check git is configured**:
+   ```bash
+   git config user.name
+   ```
+   If empty, warn: "Git user.name is not set. Run `git config --global user.name 'Your Name'` to configure it."
+
+Once all checks pass (or the user acknowledges warnings), proceed to Group 1.
+
+---
+
+### Group 1: Project Identity
+
+1. **What's the project name?** (e.g., "Immutable Audience — Mobile", "Acme Growth Strategy")
+2. **One-line description** — what is this project about in one sentence? (e.g., "Evaluate the mobile gaming market for audience monetisation opportunities")
+3. **What's the goal?** — a specific, measurable objective with a deadline (e.g., "Validate product-market fit with 20 customer interviews and a go/no-go recommendation by 30 June 2026")
+4. **Who's on the team?** — names and roles (e.g., "Dhanish (discovery), Sherry & Jeet (product)")
+
+### Group 1A: Goals & Experience
+
+14. **What's your main goal with this project?**
+    - Learning how to run a strategy project
+    - Running a real strategy project with deliverables
+    - Coordinating a team around shared strategic work
+    - Exploring and evaluating this tool
+    (e.g., "Running a real project — we need to decide whether to enter the mobile gaming market by Q3")
+
+15. **How familiar are you with strategy frameworks?**
+    - **New to this** — I haven't used formal strategy tools before
+    - **Some experience** — I've done market research or competitive analysis
+    - **Experienced** — I regularly work with strategic frameworks and hypothesis-driven methods
+
+    If "New to this": onboarding will include extra guidance in template files and suggest starting with the getting-started guide.
+    If "Experienced": standard templates, no extra guidance.
+
+### Group 2: Strategic Context
+
+5. **What's the scope?** — what's in scope and what's explicitly out of scope? (e.g., In scope: "ANZ mobile gaming studios with >$1M revenue." Out of scope: "Console gaming, studios outside ANZ")
+6. **What's the core hypothesis or strategic bet?** — the thing you're trying to prove or disprove (e.g., "Mid-tier mobile studios will pay $2k+/mo for real-time audience analytics because existing tools have a 48-hour lag")
+7. **Who's the target customer/ICP?** — describe the ideal customer profile (e.g., "Head of Growth at a mobile gaming studio, 10–50 employees, $1M–$10M ARR, currently using Firebase + manual spreadsheets")
+
+### Group 3: Project Type & Modules
+
+8. **What type of project is this?**
+   - **Market entry** (new market, new product) → all modules, entity type: "prospect"
+   - **Growth strategy** (existing product, new channels/segments) → all modules, entity type: "prospect"
+   - **Competitor research** (competitive intelligence, landscape mapping) → no discovery/pipeline, dashboard on, entity type: "competitor"
+   - **Product launch / GTM** (go-to-market planning) → all modules, entity type: "prospect"
+   - **Internal implementation** (system rollout, process change) → pipeline + dashboard, no discovery, entity type: "workstream"
+   - **Vendor / partner evaluation** (tool or partner selection) → all modules, entity type: "vendor"
+   - **Due diligence** (M&A, investment evaluation) → all modules, entity type: "target"
+   - **Business case** (investment decision, strategic proposal) → no discovery/pipeline, optional dashboard, entity type: "option"
+   - **Transformation / change** (organisational transformation) → all modules, entity type: "initiative"
+   - **Custom** → ask which modules to include
+
+   (e.g., "Internal implementation — we're rolling out a new CRM across the sales team")
+
+9. **Do you need a discovery/outreach pipeline?** (Yes/No)
+   - If yes: What's the primary entity you're tracking? (e.g., "studio", "customer", "partner", "company")
+   - If no: Will skip discovery/, pipeline, outreach-sequence skill, gtm-specialist agent, discovery-strategist agent
+
+10. **Do you have a CSV source-of-truth file yet?** (Yes/path or No)
+    - If yes: use that path for `{{PIPELINE_SOURCE_OF_TRUTH}}` (e.g., "data/gaming-studios.csv")
+    - If no: default to `data/entities.csv`
+
+### Group 3A: Feature Selection
+
+16. **Which features do you want to use?** (You can always enable more later)
+
+    | Feature | What it does | Default |
+    |---------|-------------|---------|
+    | Scoring matrix | Score and compare strategic options on custom dimensions | On |
+    | Kill conditions | Define falsifiable thresholds that signal when to stop or pivot | On |
+    | Evidence grading | Tag every claim with [CONFIRMED], [SECONDARY], [INFERENCE], or [ASSUMPTION] | On |
+    | Weekly reports | Generate periodic stakeholder summaries via `/weekly-report` | On |
+    | Context snapshots | Pre-computed summaries for fast session loading | On |
+    | Discovery pipeline | Track outreach, calls, and entity status | Based on Q9 |
+    | Dashboard | Live web dashboard deployed to Vercel | Based on Q11 |
+
+    Ask: "Are there any features you'd like to turn OFF? (Enter numbers, or 'all on' to keep defaults)"
+    (e.g., "Turn off scoring and weekly reports — we're just doing research for now")
+
+### Group 4: Dashboard & Deployment
+
+11. **Do you want a live dashboard?** (Yes/No)
+    - If yes: What Vercel project name or URL? → sets `{{DASHBOARD_URL}}` (e.g., "https://acme-strategy.vercel.app")
+    - If no: Dashboard files can be deleted
+
+### Group 5: Scoring Dimensions
+
+12. **What dimensions do you want to score your strategic options on?** (e.g., White Space, Urgency, Feasibility, Defensibility, Revenue Potential)
+    - Suggest defaults based on project type:
+      - Market entry: White Space, Urgency, Feasibility, Defensibility, Revenue Potential
+      - Growth: Growth Potential, CAC Efficiency, Retention Impact, Speed to Market, Scalability
+      - Competitor research: Market Share, Product Strength, Pricing Power, Growth Trajectory, Threat Level
+      - Product launch / GTM: Market Readiness, Channel Fit, Competitive Timing, Resource Requirement, Revenue Impact
+      - Internal implementation: Business Impact, Technical Complexity, Change Readiness, Resource Cost, Timeline Risk
+      - Vendor / partner evaluation: Capability Fit, Total Cost, Integration Effort, Vendor Stability, Lock-in Risk
+      - Due diligence: Strategic Fit, Financial Health, Integration Risk, Synergy Potential, Cultural Alignment
+      - Business case: ROI Potential, Strategic Alignment, Execution Risk, Stakeholder Support, Opportunity Cost
+      - Transformation / change: Impact Scope, Organisational Readiness, Resource Requirement, Risk Level, Time to Value
+    - Update `memory/scoring.md` with chosen dimensions
+    - Example: "We use: Market Size, Willingness to Pay, Technical Feasibility, Speed to Launch, Defensibility"
+
+### Group 6: Kill Conditions
+
+13. **Define 4–6 kill conditions** — falsifiable thresholds that, if crossed, mean the project should stop or pivot. Each should be specific and measurable.
+    - Provide examples based on project type:
+      - Market entry: "<4/15 prospects rank pain in top 3", "No WTP signal above $12k/yr after 10 calls"
+      - Growth: "CAC exceeds $120 after 3 months", "Conversion rate below 2% after 500 trials"
+      - Competitor research: "Fewer than 3 viable competitors identified after 4 weeks", "No clear white space found in the landscape"
+      - Product launch / GTM: "Pre-launch signups below 500 after 6 weeks", "Channel partner conversion below 5%"
+      - Internal implementation: "User adoption below 40% after 3 months", "Integration costs exceed budget by >30%"
+      - Vendor / partner evaluation: "No vendor scores above 3/5 on critical requirements", "Total cost exceeds budget ceiling"
+      - Due diligence: "Revenue claims can't be verified from available data", "Cultural alignment scores below threshold after site visits"
+      - Business case: "Board rejects proposal at Q2 review", "ROI falls below 15% in revised model"
+      - Transformation / change: "Key sponsor leaves the organisation", "Employee engagement scores drop >10 points"
+    - Example answer: "1. Fewer than 3/10 studios rank our pain point in their top 3. 2. No willingness-to-pay signal above $1k/mo after 15 discovery calls. 3. TAM estimate falls below $50M. 4. Unable to secure 5 design partners in 6 weeks."
+
+---
+
+## Validation / Dry-Run Step
+
+After all answers are collected (whether via Quick Start or Full Setup) but **before** any file replacement, present a summary table to the user for confirmation:
+
+> **Here's what I'm about to set. Please confirm or request changes.**
+>
+> | Placeholder | Value |
+> |---|---|
+> | `{{PROJECT_NAME}}` | _answer_ |
+> | `{{PROJECT_SLUG}}` | _auto-generated_ |
+> | `{{ONE_LINE_DESCRIPTION}}` | _answer_ |
+> | `{{GOAL}}` | _answer_ |
+> | `{{TEAM}}` | _answer_ |
+> | `{{SCOPE_DEFINITION}}` | _answer_ |
+> | `{{OUT_OF_SCOPE}}` | _answer_ |
+> | `{{STRATEGIC_HYPOTHESIS}}` | _answer_ |
+> | `{{ICP_DESCRIPTION}}` | _answer_ |
+> | `{{ENTITY_TYPE}}` | _answer_ |
+> | `{{ENTITY_TYPE_PLURAL}}` | _auto-generated_ |
+> | `{{PIPELINE_SOURCE_OF_TRUTH}}` | _answer or default_ |
+> | `{{DASHBOARD_URL}}` | _answer or default_ |
+> | `{{DIM_1}}` – `{{DIM_6}}` | _listed dimensions_ |
+> | `{{DATE}}` | _today's date_ |
+> | **Modules included** | discovery: yes/no, dashboard: yes/no |
+> | **Kill conditions** | _listed_ |
+> | **Goal** | _{answer from Q14 or Quick Start Q6}_ |
+> | **Experience level** | _{answer from Q15, or "Not asked" if Quick Start}_ |
+> | **Features disabled** | _{list of disabled features, or "None"}_ |
+>
+> **Proceed?** (Yes / Edit a value / Start over)
+
+Only proceed with file modifications once the user confirms.
+
+---
+
+## After Questions — Bulk Replace
+
+Once the user confirms and all answers are finalised:
+
+1. **Search and replace all `{{PLACEHOLDER}}` values** across every file in the project:
+   - `{{PROJECT_NAME}}` → answer to Q1
+   - `{{PROJECT_SLUG}}` → lowercase, hyphenated version of Q1
+   - `{{ONE_LINE_DESCRIPTION}}` → Q2
+   - `{{GOAL}}` → Q3
+   - `{{TEAM}}` → Q4
+   - `{{SCOPE_DEFINITION}}` → Q5 (in-scope part)
+   - `{{OUT_OF_SCOPE}}` → Q5 (out-of-scope part)
+   - `{{STRATEGIC_HYPOTHESIS}}` → Q6
+   - `{{ICP_DESCRIPTION}}` → Q7
+   - `{{ENTITY_TYPE}}` → Q9 entity name (singular)
+   - `{{ENTITY_TYPE_PLURAL}}` → Q9 entity name (plural)
+   - `{{PIPELINE_SOURCE_OF_TRUTH}}` → Q10
+   - `{{DASHBOARD_URL}}` → Q11
+   - `{{DIM_1}}` through `{{DIM_6}}` → Q12 dimensions
+   - `{{DATE}}` → today's date
+
+2. **Populate kill conditions** in:
+   - `memory/MEMORY.md` Kill Conditions table
+   - `docs/executive-summary.md` §6 Kill Conditions table
+   - `memory/discovery.md` Kill Condition Tracker (if discovery module active)
+   - `context/project-state.md` Kill Conditions table
+
+3. **Remove optional modules** based on project type (Q8) and user answers (Q9):
+
+   Module defaults by project type:
+   - Market entry → discovery: yes, pipeline: yes, dashboard: yes
+   - Growth strategy → discovery: yes, pipeline: yes, dashboard: yes
+   - Competitor research → discovery: no, pipeline: no, dashboard: yes
+   - Product launch / GTM → discovery: yes, pipeline: yes, dashboard: yes
+   - Internal implementation → discovery: no, pipeline: yes, dashboard: yes
+   - Vendor / partner evaluation → discovery: yes, pipeline: yes, dashboard: yes
+   - Due diligence → discovery: yes, pipeline: yes, dashboard: yes
+   - Business case → discovery: no, pipeline: no, dashboard: optional
+   - Transformation / change → discovery: yes, pipeline: yes, dashboard: yes
+   - Custom → ask user
+
+   Apply removals:
+   - If no discovery: delete `discovery/`, `memory/discovery.md`, `context/pipeline-state.md`, `dashboard/pipeline.html`, `skills/pipeline-update/`, `skills/outreach-sequence/`, `.claude/agents/discovery-strategist.md`, `.claude/agents/gtm-specialist.md`
+   - If no pipeline: delete `skills/pipeline-update/`, `skills/outreach-sequence/`, `.claude/agents/gtm-specialist.md`
+   - If no dashboard: delete `dashboard/`
+
+4. **Apply feature-level removals** based on Q16 (Full Setup) or structure selection (Quick Start):
+
+   - **Scoring off**: Skip Group 5, remove `memory/scoring.md`, remove `dashboard/scoring.html`, skip `/compare-options` skill mention in CLAUDE.md skills table
+   - **Kill conditions off**: Skip Group 6, remove kill condition sections from template files (leave the table headers but mark as "Not configured — enable kill conditions in project.config.json to use this feature")
+   - **Evidence grading off**: Add a note to CLAUDE.md: "Evidence grading is disabled for this project. Claims are not tagged with confidence grades."
+   - **Weekly reports off**: Remove `/weekly-report` skill reference
+   - **Context snapshots off**: Remove `context/` directory, update CLAUDE.md to note snapshots are disabled
+   - **Minimal mode**: Removes `discovery/`, `dashboard/`, pipeline skills, scoring, kill conditions, evidence grading, weekly reports, context snapshots — keeps only `research/`, `memory/` (MEMORY.md, research.md, decisions.md), `docs/`, and basic skills (onboard, session-start, session-end, health-check, enrich-entity, critical-reasoning, decision, synthesise)
+
+5. **Remove remaining `{{PLACEHOLDER}}` markers** — scan all files for any unreplaced `{{` patterns and either fill them with sensible defaults or flag them for the user.
+
+### Step 5.5: Generate Sample Entity CSV
+
+If the discovery module is active and no existing CSV was provided (Q10 = No), create `data/entities.csv` with the correct headers and 2 example rows so the user can see the expected format:
+
+```csv
+name,slug,status,tier,category,last_contact,next_action,channel,notes
+Example Corp,example-corp,Not started,Tier 2,Direct competitor,,Research website,Web,Added as example — delete this row
+Demo Inc,demo-inc,Not started,Tier 3,Adjacent,,Initial research,Web,Added as example — delete this row
+```
+
+Ensure the `data/` directory exists before writing the file.
+
+### Step 5.6: Generate project.config.json
+
+Write the project configuration to `project.config.json` at the repo root using the collected answers. Follow this schema:
+
+```json
+{
+  "templateVersion": "1.0.0",
+  "templateSource": "github.com/DiffTheEnder/DSS-Claude-Stack",
+  "projectName": "<Q1>",
+  "projectSlug": "<auto-generated slug>",
+  "projectType": "{selected type}",
+  "oneLineDescription": "<Q2>",
+  "goal": "<Q3>",
+  "team": "<Q4>",
+  "scope": "<Q5 in-scope>",
+  "outOfScope": "<Q5 out-of-scope>",
+  "strategicHypothesis": "<Q6>",
+  "icpDescription": "<Q7>",
+  "entityType": "<Q9 singular>",
+  "entityTypePlural": "<Q9 plural>",
+  "pipelineSourceOfTruth": "<Q10 or default>",
+  "dashboardUrl": "<Q11 or default>",
+  "modules": {
+    "discovery": true/false,
+    "pipeline": true/false,
+    "dashboard": true/false
+  },
+  "features": {
+    "scoring": true/false,
+    "killConditions": true/false,
+    "evidenceGrading": true/false,
+    "weeklyReports": true/false,
+    "contextSnapshots": true/false
+  },
+  "scoringDimensions": ["<Q12 dimensions>"],
+  "killConditions": ["<Q13 conditions>"]
+}
+```
+
+6. **Run verification**:
+   ```bash
+   grep -r "{{" --include="*.md" --include="*.js" --include="*.html" .
+   ```
+   Should return zero results (all placeholders replaced).
+
+### Step 6.5: Transform README
+
+After verification, replace the open-source template README with a project-instance README. Generate a new `README.md` at the repo root containing:
+
+- **Project name** as the top-level heading
+- **Description** — the one-liner from Q2
+- **Goal** — the measurable objective from Q3
+- **Team** — from Q4
+- **Status** — "Initialised — onboarding complete, discovery not yet started"
+- **Folder map** — a brief overview of the directory structure and what each folder contains
+- **Key docs** — links to:
+  - `memory/MEMORY.md` (project memory)
+  - `docs/executive-summary.md` (executive summary)
+  - `context/project-state.md` (current state)
+  - `memory/scoring.md` (scoring framework)
+  - `memory/discovery.md` (discovery tracker, if active)
+- **Dashboard** — link to `{{DASHBOARD_URL}}` if the dashboard module is active, otherwise note "Dashboard not configured"
+- **Generated by** — "Initialised from DS Strategy Stack on `{{DATE}}`"
+
+Do **not** include template-level instructions, contribution guidelines, or licence information from the original README — this is now a project-specific document.
+
+7. **Rebuild dashboard** (if kept):
+   ```bash
+   cd dashboard && npm install && node build-data.js
+   ```
+
+8. **Report completion** to the user with:
+   - Summary of all values set
+   - List of modules included/excluded
+   - List of features enabled/disabled
+   - Files generated (entity CSV, project.config.json, README)
+   - Suggested first actions (e.g., "Start with competitor research", "Define your first 10 targets")
+
+---
+
+## After Onboarding
+
+This skill is no longer needed. The user can:
+- Delete the `skills/onboard/` directory
+- Or keep it for reference
+
+Suggest: "Onboarding complete. You can delete `skills/onboard/` — it's no longer needed."

package/template/skills/outreach-sequence/SKILL.md

@@ -0,0 +1,79 @@
+---
+name: outreach-sequence
+description: Use when designing or reviewing an outreach cadence for a set of {{ENTITY_TYPE_PLURAL}}, or when the user asks "what should I send next to X". Produces a sequenced, personalised multi-touch plan with draft messages.
+---
+
+# Outreach Sequence
+
+**Flexible skill** — adapt the structure to the project context and {{ENTITY_TYPE}}.
+
+---
+
+## Sequence Types
+
+| Type | When | Touches | Spacing |
+|------|------|---------|---------|
+| Cold | No prior relationship | 3–4 | 3–5 days |
+| Warm intro follow-up | Mutual connection made intro | 2–3 | 2–3 days |
+| Post-non-response | Sent outreach, no reply | 2–3 | 5–7 days |
+| Post-meeting follow-up | After a call/meeting | 2–3 | 1–3 days |
+| Re-engagement | Gone cold after prior contact | 2 | 7–14 days |
+
+---
+
+## Voice Standards
+
+All outreach must follow these rules:
+
+- **Short sentences**. No compound sentences with multiple clauses.
+- **Plain words**. No corporate filler ("leverage", "synergise", "circle back", "deep dive").
+- **Under 100 words** per message. Shorter is better.
+- **One ask per message**. Never combine "can we chat?" with "also could you share X?"
+- **Personalised opening**. Reference something specific to the {{ENTITY_TYPE}} — a recent event, their product, a shared connection.
+- **Clear next step**. Every message ends with a specific, easy-to-action request.
+- **No attachments on first touch**. Earn attention before sending materials.
+
+---
+
+## Personalisation Rules
+
+For each {{ENTITY_TYPE}}, personalise based on:
+
+1. **Their recent activity** — product launches, funding, hiring, press mentions
+2. **Their specific pain** — if known from research, reference it directly
+3. **Mutual connections** — if any exist, lead with the introduction
+4. **Their role** — adapt language for C-level vs practitioner vs technical buyer
+5. **Channel norms** — email is more formal; LinkedIn is more conversational
+
+---
+
+## Output Format
+
+For each {{ENTITY_TYPE}}, produce:
+
+```markdown
+## {{ENTITY_NAME}} — {{SEQUENCE_TYPE}} Sequence
+
+**Channel**: Email / LinkedIn / Intro request
+**Personalisation hook**: [what makes this outreach specific to them]
+
+### Touch 1 (Day 0)
+Subject: [if email]
+Body: [draft message]
+
+### Touch 2 (Day X)
+Subject: [if email]
+Body: [draft message]
+
+### Touch 3 (Day X) — optional
+Subject: [if email]
+Body: [draft message]
+```
+
+---
+
+## After Drafting
+
+1. Update pipeline tracker with outreach status
+2. Update `memory/discovery.md` Outreach Log
+3. If outreach requires call prep, create `discovery/prep/{{ENTITY_SLUG}}-prep.md`

package/template/skills/pipeline-update/SKILL.md

@@ -0,0 +1,90 @@
+---
+name: pipeline-update
+description: Use when the user reports a pipeline status change — intro received, meeting booked, outreach sent, response received, entity goes cold or dead. Executes all pipeline update steps in order without manual prompting.
+---
+
+# Pipeline Update
+
+**Rigid skill** — execute ALL steps in order. Do not skip steps or ask whether to proceed.
+
+---
+
+## Status Lifecycle
+
+| From | To | Trigger |
+|------|----|---------|
+| (new) | Not started | Entity added to pipeline |
+| Not started | Contacted | Outreach sent (any channel) |
+| Not started | Engaged | Inbound interest or warm intro |
+| Contacted | Engaged | Response received |
+| Engaged | Meeting booked | Call/meeting scheduled |
+| Meeting booked | Meeting completed | Call/meeting happened |
+| Any | Dead | Disqualified, unresponsive, or explicitly declined |
+
+### Alternative Lifecycle (for implementation, transformation, and internal projects)
+
+| From | To | Trigger |
+|------|----|---------|
+| (new) | Not started | Entity/workstream added to pipeline |
+| Not started | In progress | Work has begun |
+| In progress | Blocked | Dependency or issue preventing progress |
+| Blocked | In progress | Blocker resolved |
+| In progress | Completed | Work finished and verified |
+| Any | Cancelled | Workstream deprioritised or abandoned |
+
+Use this lifecycle when the project type is **Internal Implementation**, **Transformation / Change**, or any project where entities represent workstreams or initiatives rather than outreach targets.
+
+The project type is stored in `project.config.json` under `projectType`. Check this to determine which lifecycle to use.
+
+---
+
+## Step 0 — Identify the Event
+
+What happened? Map it to a status transition from the table above. If ambiguous, ask the user.
+
+Check `project.config.json` `projectType` to determine which status lifecycle to use (outreach vs implementation).
+
+## Step 1 — Check if New Entity
+
+If this {{ENTITY_TYPE}} doesn't exist in the source of truth yet:
+- Add to `{{PIPELINE_SOURCE_OF_TRUTH}}` (CSV is source of truth — never add only to JSON)
+- Fill all available columns
+
+## Step 2 — Update Pipeline Tracker
+
+Update the tracker data with:
+- New status
+- Last contact date
+- Next action and due date
+- Channel used
+- Any notes
+
+## Step 3 — Update Memory
+
+Update `memory/discovery.md` Outreach Log with:
+- Date, {{ENTITY_TYPE}}, channel, action, result, next step
+
+Update `memory/MEMORY.md` Discovery line if the count or status summary has changed.
+
+## Step 4 — Create Prep File (if meeting booked)
+
+If status → Meeting booked:
+- Create `discovery/prep/{{ENTITY_SLUG}}-prep.md` using `templates/call-prep.md`
+- Research the {{ENTITY_TYPE}} before filling the prep template
+
+## Step 5 — Handle Dead Status
+
+If status → Dead:
+- Note reason in memory/discovery.md Outreach Log
+- Update MEMORY.md to remove from active lists
+- No prep file needed
+
+## Step 6 — Rebuild Dashboard
+
+```bash
+cd dashboard && node build-data.js
+```
+
+## Step 7 — Confirm to User
+
+Report: what changed, what files were updated, what the next action is.