create-dss-project 0.1.0

package/template/skills/process-call/SKILL.md

@@ -0,0 +1,96 @@
+---
+name: process-call
+description: Use after a discovery or stakeholder call to process raw notes into structured files. Extracts pain points, WTP signals, hypothesis evidence, and updates all memory files and the dashboard. Rigid skill — execute all steps.
+---
+
+# Process Call
+
+**Rigid skill** — execute ALL steps in order. Do not skip steps or ask whether to proceed.
+
+Use this after any discovery, intel, or stakeholder call to transform raw notes into structured project data.
+
+---
+
+## Step 0 — Gather Raw Input
+
+Ask the user for:
+1. **Entity name** — who was the call with?
+2. **Call type** — Discovery / Intel / Follow-up / Partnership
+3. **Attendees** — who was on the call?
+4. **Raw notes** — pasted text, a file path, or dictated summary
+
+If the user pastes raw notes, work with those. If they provide a file path, read that file.
+
+## Step 1 — Create Structured Call Notes
+
+Create `discovery/calls/{entity-slug}-{YYYY-MM-DD}.md` using the template at `templates/call-notes.md`.
+
+Extract from the raw notes and populate:
+- **Key Takeaways**: 3–5 bullet points — the most important things learned
+- **Pain Point Rankings**: intensity 1–5, with direct quotes where available
+- **WTP / Value Signals**: any willingness-to-pay indicators with context and evidence grade
+- **Key Quotes**: verbatim quotes that capture insights (attribute to speaker)
+- **Hypotheses Tested**: map findings against the project's strategic hypothesis and kill conditions
+- **Learnings & Surprises**: what challenged assumptions or was unexpected
+- **Next Steps**: concrete actions with owners and due dates
+- **Impact Assessment**: which kill conditions, hypotheses, or research gaps are affected
+
+Every factual claim must be tagged: [CONFIRMED], [SECONDARY], [INFERENCE], or [ASSUMPTION].
+
+## Step 2 — Update Pain Point Rankings
+
+Read `memory/discovery.md`. Update the Pain Point Rankings table:
+- If a pain point from this call already exists → update frequency count and intensity if this call provides stronger evidence
+- If a pain point is new → add it to the table
+- Maintain ranking by frequency × intensity
+
+## Step 3 — Update WTP Signals
+
+In `memory/discovery.md`, update the WTP Signals table:
+- Add any new signals from this call
+- Include entity name, signal, context, and evidence grade
+
+## Step 4 — Update Kill Condition Tracker
+
+In `memory/discovery.md`, update the Kill Condition Tracker:
+- For each kill condition, check if this call provided new evidence
+- Update the evidence column and status (UNTESTED → PASSING or FAILING)
+- If a kill condition status changes → flag this prominently to the user
+
+Also update `memory/MEMORY.md` Kill Conditions table if any status changed.
+
+## Step 5 — Update Pipeline Status
+
+Update the entity's status in the pipeline:
+- If they were "Meeting booked" → move to "Meeting completed"
+- Update last contact date, next action, and notes in the pipeline tracker
+- Update `memory/discovery.md` Outreach Log with call details
+- Update `memory/MEMORY.md` Discovery line (call count, status summary)
+
+## Step 6 — Hypothesis Impact Check
+
+Read the project's strategic hypothesis from `memory/MEMORY.md`.
+
+Does this call provide evidence that **supports**, **weakens**, or is **neutral** to the hypothesis?
+
+- If **supports or weakens** → run `/critical-reasoning` on the updated evidence before updating `docs/executive-summary.md`
+- If **neutral** → note in the call notes Impact Assessment and proceed
+
+## Step 7 — Rebuild Dashboard
+
+```bash
+cd dashboard && node build-data.js
+```
+
+## Step 8 — Report to User
+
+Provide a structured summary:
+
+> **Call processed — {entity name} ({date})**
+> - **Key insight**: {single most important takeaway}
+> - **Pain points**: {count} identified ({count} new)
+> - **WTP signals**: {count} captured
+> - **Kill conditions affected**: {list or "none"}
+> - **Hypothesis impact**: Supports / Weakens / Neutral
+> - **Next actions**: {list from call notes}
+> - **Files updated**: {list}

package/template/skills/rebuild-snapshots/SKILL.md

@@ -0,0 +1,88 @@
+---
+name: rebuild-snapshots
+description: Regenerate all 4 context snapshots from raw source files. Run when snapshots have drifted from reality — after multiple sessions, bulk research, or when /session-start reports stale data.
+---
+
+# Rebuild Snapshots
+
+Regenerate the pre-computed context snapshots in `context/` from raw source files. These snapshots enable fast context loading at session start — but they drift when source files change without snapshot updates.
+
+---
+
+## When to Run
+
+- `/session-start` reports data is >7 days old
+- After bulk research or multiple discovery calls
+- After strategic decisions that change the project direction
+- When `/session-end` detects relevant source file changes
+- Manually, whenever snapshots feel out of date
+
+---
+
+## Snapshot 1 — Project State (`context/project-state.md`)
+
+**Read these source files:**
+- `memory/MEMORY.md` — project state, hypothesis, kill conditions, priorities
+- `memory/decisions.md` — recent strategic decisions
+- `memory/discovery.md` — kill condition tracker, discovery progress
+- `docs/executive-summary.md` — current hypothesis framing, risk register
+
+**Regenerate with these sections:**
+1. **Mission & Hypothesis** — one-paragraph summary of what we're doing and why
+2. **Kill Conditions** — current status table (copy from MEMORY.md, enrich with evidence from discovery.md)
+3. **Discovery Progress** — calls completed, pipeline summary, key patterns emerging
+4. **Recent Decisions** — last 3–5 decisions from decisions.md
+5. **Current Priorities** — from MEMORY.md
+6. **Top Risks** — from executive summary risk register
+
+---
+
+## Snapshot 2 — Market Snapshot (`context/market-snapshot.md`)
+
+**Read these source files:**
+- `memory/research.md` — market sizing summary
+- `research/market/` — all market research files
+- `docs/executive-summary.md` §2 (Market Opportunity)
+
+**Regenerate with these sections:**
+1. **TAM / SAM / SOM** — current estimates with sources and evidence grades
+2. **Market Dynamics** — tailwinds and headwinds
+3. **Key Trends** — 3–5 trends shaping the market
+4. **Timing Window** — convergence or urgency signals
+
+---
+
+## Snapshot 3 — Pipeline State (`context/pipeline-state.md`)
+
+**Read these source files:**
+- Pipeline source of truth CSV (path from `project.config.json` or `memory/MEMORY.md`)
+- `memory/discovery.md` — outreach log
+- `memory/MEMORY.md` — discovery line
+
+**Regenerate with these sections:**
+1. **Pipeline Summary** — total entities, breakdown by status
+2. **Status Breakdown** — table of all entities with current status, last contact, next action
+3. **Overdue Actions** — any next actions with dates in the past
+4. **Recent Activity** — last 5 pipeline events from outreach log
+
+---
+
+## Snapshot 4 — Competitor Snapshot (`context/competitor-snapshot.md`)
+
+**Read these source files:**
+- `memory/research.md` — entity capability map
+- `research/competitors/` — all competitor research files (read TL;DR sections at minimum)
+
+**Regenerate with these sections:**
+1. **Competitive Landscape** — capability map table (copy from research.md)
+2. **Key Patterns** — 3–5 patterns across competitors (common strengths, shared gaps)
+3. **White Space** — opportunities not covered by existing players
+4. **Convergence Risks** — competitors moving towards our intended position
+
+---
+
+## After Rebuilding
+
+1. Add today's date as "Last regenerated" at the top of each snapshot
+2. Report which snapshots changed meaningfully vs. were already current
+3. If any snapshot reveals contradictions with `memory/MEMORY.md` → flag to the user

package/template/skills/session-end/SKILL.md

@@ -0,0 +1,120 @@
+---
+name: session-end
+description: Run at the end of every work session. Executes the full 10-step housekeeping checklist — file audit, memory updates, hypothesis review, status sync, dashboard rebuild, and commit. Rigid skill — execute all steps in order.
+---
+
+# Session End
+
+**Rigid skill** — execute ALL steps in order. Do not skip steps or ask whether to proceed.
+
+Run this before finishing any session that produced or modified files.
+
+---
+
+## Step 1 — File Placement Audit
+
+Scan all files created or modified this session. Verify each is in the correct location:
+
+| File type | Correct location | Template |
+|-----------|-----------------|----------|
+| Entity teardown | `research/competitors/{slug}.md` | `templates/entity-teardown.md` |
+| Market/TAM research | `research/market/` | — |
+| Technical research | `research/technical/` | — |
+| Call notes | `discovery/calls/{entity}-{date}.md` | `templates/call-notes.md` |
+| Call prep | `discovery/prep/{entity}-prep.md` | `templates/call-prep.md` |
+| Strategy memo | `docs/memos/` | — |
+| Deck/presentation | `docs/decks/` | — |
+| Data file (CSV/JSON) | `data/` | — |
+
+If a file is in the wrong place → move it now. If created without the correct template → flag it.
+
+## Step 2 — Memory Cascade Update
+
+Check which work was done this session and update the corresponding memory files:
+
+| Work done | Update required |
+|-----------|----------------|
+| Competitor/entity research | Add/update row in `memory/research.md` |
+| New target entities found | Update Target Entity List in `memory/research.md` |
+| Technical pathway research | Update Technical Findings in `memory/research.md` |
+| Outreach sent (any channel) | Update Outreach Log in `memory/discovery.md` + `memory/MEMORY.md` Discovery line |
+| Call prep created | Note in pipeline tracker for the matching entity |
+| Call notes processed | Note in pipeline tracker for the matching entity |
+| Entity disqualified | Update `memory/MEMORY.md` Discovery line + note reason in `memory/discovery.md` |
+| Customer discovery call | Update Pain Point Rankings + WTP Signals in `memory/discovery.md` + `memory/MEMORY.md` |
+| Kill condition evidence changed | Update Kill Condition Tracker in `memory/discovery.md` + `memory/MEMORY.md` |
+| Strategic decision made | Add entry to `memory/decisions.md` + one-liner to `memory/MEMORY.md` |
+| Option scoring changed | Update `memory/scoring.md` |
+
+## Step 3 — Hypothesis Review
+
+Ask: "Did this session produce new evidence that could affect hypothesis confidence?"
+
+- If **YES** → run `/critical-reasoning` before updating the executive summary. Only update `docs/executive-summary.md` after the user confirms the briefing.
+- If **NO** (structural/setup work) → proceed to Step 4.
+
+## Step 4 — Status Blurb Update
+
+Update `docs/output/status-blurb.md`. Rules:
+- Max 4 sentences
+- Plain English — no jargon, no technical codes, no internal terminology
+- Never mention internal tooling, agent workflows, file reorganisation, or process changes
+- Cover: current phase, what's active, key unknown, any blockers
+
+## Step 5 — README Status Sync
+
+Copy the content of `docs/output/status-blurb.md` into the `<!-- STATUS:START -->` / `<!-- STATUS:END -->` block in `README.md`.
+
+## Step 6 — Project Activity Update
+
+Apply the filter — only log exec-worthy progress:
+- **Log**: Concrete deliverables, milestones, external-facing actions (outreach sent, calls completed, research synthesis, memos)
+- **Skip**: Dashboard UI work, CSS fixes, skill creation, data schema changes, call prep creation, internal tooling, maintenance, housekeeping
+- **Rule of thumb**: If you wouldn't mention it in a 2-minute standup with an investor, don't log it
+
+If there is exec-worthy progress → append to `docs/output/project-activity.md`.
+
+## Step 7 — STATUS.md & Work Log Update
+
+All of the following — never do (a) without (b):
+
+a) Move completed "In Progress" rows → STATUS.md "Completed Recently"
+b) Append the SAME rows to `docs/output/work-log.md` (append-only, never trim)
+c) Trim STATUS.md "Completed Recently" entries older than 1 week (safe — they're in work-log.md)
+d) Remove "Blocked / Waiting" entries that are resolved
+
+## Step 8 — Rebuild Context Snapshots
+
+Run `/rebuild-snapshots` if any of the following changed this session:
+- Discovery calls processed
+- Kill conditions updated
+- Strategic decisions made
+- Market research added
+- Pipeline state changed significantly
+
+If none of these changed → skip this step.
+
+## Step 9 — Rebuild Dashboard
+
+```bash
+cd dashboard && node build-data.js
+```
+
+Include regenerated `dashboard/data/*.json` files in the commit.
+
+## Step 10 — Commit & Push
+
+1. Stage all files modified during the session and housekeeping (use specific file paths, never `git add -A`)
+2. Commit with a concise message summarising the session's work (not the housekeeping itself)
+3. Push to remote
+
+## Step 11 — Report
+
+Report to the user:
+- Files created/modified
+- Memory files updated
+- Whether hypothesis was reviewed
+- Dashboard rebuild status
+- Commit hash
+
+> **Session complete.** {1-sentence summary of what was accomplished.}

package/template/skills/session-start/SKILL.md

@@ -0,0 +1,93 @@
+---
+name: session-start
+description: Run at the beginning of every work session. Loads context, checks for conflicts, shows priorities and overdue actions. Ensures you start every session oriented and ready to work.
+---
+
+# Session Start
+
+**Run this at the start of every session** — it orients you to the current project state and prevents duplicate or conflicting work.
+
+---
+
+## Step 0 — Get Current Date
+
+```bash
+date +%Y-%m-%d
+```
+
+Record this as the session date.
+
+## Step 1 — Check for Conflicts
+
+Read `STATUS.md`. Check the "In Progress" table:
+- If another agent has items in progress → flag to the user which workstreams are active
+- If the user's intended work overlaps with an in-progress item → warn before proceeding
+
+## Step 2 — Determine Context Loading Mode
+
+Ask the user: **"What are you working on this session?"**
+
+Map their answer to a loading mode:
+
+| Task type | Mode | Load |
+|-----------|------|------|
+| Pipeline updates, outreach, scheduling, call prep | **Fast** | `context/pipeline-state.md` only |
+| Research, competitor analysis, entity enrichment | **Standard** | `context/competitor-snapshot.md` + `context/market-snapshot.md` |
+| Discovery call prep, hypothesis work, strategic review | **Standard** | `context/project-state.md` |
+| Hypothesis review, critical reasoning, kill condition updates, strategic decisions | **Deep** | `context/project-state.md` → then ALL raw evidence files |
+| Housekeeping, maintenance | **Fast** | `context/project-state.md` |
+
+Tell the user which mode you're using and why.
+
+## Step 3 — Load Context Snapshot(s)
+
+Read the appropriate context snapshot(s) based on Step 2.
+
+For **Deep** mode: read the snapshot first for orientation, then read:
+- All files in `discovery/calls/`
+- `memory/decisions.md`
+- `memory/research.md`
+
+## Step 4 — Show Current Priorities
+
+Read `memory/MEMORY.md`. Report to the user:
+- **Current phase** and project state
+- **Current priority** (the 1–2 items flagged as most important)
+- **Kill condition status** — any conditions that are FAILING or changed recently
+
+## Step 5 — Check for Overdue Actions
+
+If the discovery module is active, scan `memory/discovery.md` Outreach Log and `context/pipeline-state.md` for:
+- Next actions with dates in the past
+- Entities in "Meeting booked" status with no follow-up
+- Entities in "Contacted" status for more than 7 days with no response
+
+Report any overdue items to the user.
+
+## Step 6 — Report Time Since Last Session
+
+Read `docs/output/work-log.md`. Find the most recent entry date. Calculate and report how many days since the last session.
+
+If > 7 days: suggest running `/rebuild-snapshots` to ensure context is fresh.
+
+## Step 7 — Register in STATUS.md
+
+Add the user's intended work to the "In Progress" table in `STATUS.md`:
+- Workstream: based on task type
+- Item: brief description of planned work
+- Owner: user's name (from `memory/MEMORY.md` Team field)
+- Date: today's date
+
+## Step 8 — Session Briefing
+
+Provide a concise briefing:
+
+> **Session briefing — {date}**
+> - **Mode**: {Fast/Standard/Deep}
+> - **Focus**: {what the user said they're working on}
+> - **Priority**: {from MEMORY.md}
+> - **Overdue**: {count of overdue items, or "none"}
+> - **Days since last session**: {N}
+> - **Conflicts**: {none, or list}
+
+Then: "Ready to start. What's first?"

package/template/skills/synthesise/SKILL.md

@@ -0,0 +1,108 @@
+---
+name: synthesise
+description: Synthesise research across multiple raw files into a structured memo with evidence grading. Use when you need to pull together competitor patterns, market dynamics, discovery themes, or any cross-cutting analysis.
+---
+
+# Synthesise
+
+Transform scattered raw research into a structured, evidence-graded synthesis memo.
+
+---
+
+## Step 0 — Define the Synthesis Topic
+
+Ask the user: **"What do you want to synthesise?"**
+
+Common synthesis types:
+
+| Type | Source files | Output |
+|------|------------|--------|
+| Competitor landscape | `research/competitors/*.md` + `memory/research.md` | Patterns, gaps, white space |
+| Market dynamics | `research/market/*.md` + `memory/research.md` | Trends, sizing, timing |
+| Discovery patterns | `discovery/calls/*.md` + `memory/discovery.md` | Pain themes, WTP signals, segments |
+| Technical feasibility | `research/technical/*.md` | Architecture options, trade-offs |
+| Custom topic | User-specified files | User-specified focus |
+
+## Step 1 — Gather All Source Material
+
+Read ALL relevant raw files — not just snapshots. Synthesis requires the full evidence base.
+
+For each file, extract:
+- Key claims and findings
+- Evidence grades on each claim
+- Contradictions or tensions between sources
+- Gaps — what's missing?
+
+## Step 2 — Identify Patterns
+
+Look for:
+- **Convergence**: 3+ sources agreeing on the same point → strong signal
+- **Contradictions**: Sources disagreeing → flag explicitly, don't silently resolve
+- **Gaps**: Important questions with no evidence → flag as research needs
+- **Surprises**: Findings that challenge the project hypothesis
+
+## Step 3 — Write the Synthesis Memo
+
+Create `docs/memos/synthesis-{topic-slug}-{date}.md` with this structure:
+
+```
+# Synthesis — {Topic}
+
+**Date**: {date}
+**Sources**: {count} files analysed
+**Analyst**: {from MEMORY.md team}
+
+## TL;DR
+
+<!-- 3-5 sentence executive summary of findings -->
+
+## Key Findings
+
+### Finding 1: {title}
+{2-3 sentences}. [CONFIRMED/SECONDARY/INFERENCE]
+
+Evidence:
+- {source 1}: {what it says} [grade]
+- {source 2}: {what it says} [grade]
+
+### Finding 2: ...
+
+## Contradictions & Tensions
+
+| Claim A | Source | Claim B | Source | Resolution |
+|---------|--------|---------|--------|------------|
+
+## Gaps & Research Needs
+
+- {gap description} — suggested next step
+- ...
+
+## Implications for Strategy
+
+- {implication 1} — how this affects the hypothesis or kill conditions
+- {implication 2}
+
+## Evidence Grade Distribution
+
+- CONFIRMED: {count}
+- SECONDARY: {count}
+- INFERENCE: {count}
+- ASSUMPTION: {count}
+
+Coverage: {tagged claims / total claims}%
+```
+
+## Step 4 — Update Memory
+
+If the synthesis reveals new patterns or changes understanding:
+- Update the relevant section of `memory/research.md` or `memory/discovery.md`
+- If it affects the hypothesis → flag for `/critical-reasoning`
+
+## Step 5 — Report
+
+> **Synthesis complete — {topic}**
+> - {count} sources analysed
+> - {count} key findings
+> - {count} contradictions flagged
+> - {count} research gaps identified
+> - Output: `docs/memos/synthesis-{slug}-{date}.md`

package/template/skills/weekly-report/SKILL.md

@@ -0,0 +1,79 @@
+---
+name: weekly-report
+description: Generate a stakeholder-ready weekly summary from project activity, work log, pipeline state, and status blurb. Outputs a formatted report and optional Slack-ready plain text version.
+---
+
+# Weekly Report
+
+Generate a periodic summary suitable for stakeholders, investors, or team standups.
+
+---
+
+## Step 1 — Gather Source Data
+
+Read the following files:
+- `docs/output/work-log.md` — filter to entries from the last 7 days (or since last report)
+- `docs/output/project-activity.md` — recent exec-worthy milestones
+- `docs/output/status-blurb.md` — current status
+- `context/pipeline-state.md` — pipeline summary
+- `memory/MEMORY.md` — kill conditions, current priority, project state
+- `memory/decisions.md` — any decisions made this week
+
+## Step 2 — Generate Report
+
+Create `docs/output/weekly-report-{YYYY-MM-DD}.md`:
+
+```
+# Weekly Report — {date range}
+
+## Status
+
+{Copy of status-blurb.md — 4 sentences max}
+
+## This Week
+
+### Completed
+- {bullet points from work-log, exec-worthy items only}
+
+### Decisions Made
+- {from decisions.md, this week only — or "No strategic decisions this week"}
+
+### Discovery & Pipeline
+- **Pipeline**: {total} entities — {breakdown by status}
+- **Calls completed**: {count this week}
+- **Key findings**: {1-2 sentence summary of discovery insights}
+
+## Kill Conditions
+
+| # | Condition | Status | Change |
+|---|-----------|--------|--------|
+{table from MEMORY.md — add "Change" column: ↑ improved, ↓ worsened, — unchanged}
+
+## Next Week
+
+### Priorities
+- {from MEMORY.md current priority}
+
+### Planned
+- {upcoming actions from pipeline next-actions}
+
+## Risks & Blockers
+
+- {any items from STATUS.md Blocked/Waiting}
+- {any kill conditions trending FAILING}
+```
+
+## Step 3 — Generate Slack Version (Optional)
+
+If the user wants a Slack-ready version, produce a plain text summary:
+- Under 500 words
+- No markdown tables (use bullet points instead)
+- Lead with the single most important update
+- End with "Priorities this week: {list}"
+
+## Step 4 — Report
+
+> **Weekly report generated — {date range}**
+> - Output: `docs/output/weekly-report-{date}.md`
+> - Period: {start} to {end}
+> - {count} items completed, {count} decisions, {count} calls

package/template/templates/call-notes.md

@@ -0,0 +1,67 @@
+# Call Notes — {{ENTITY_NAME}}
+
+**Date**: {{DATE}}
+**Attendees**: {{ATTENDEES}}
+**Call type**: Discovery / Intel / Follow-up / Partnership
+**Duration**: ~X min
+
+---
+
+## Key Takeaways
+
+<!-- 3–5 bullet points: the most important things learned -->
+
+1.
+2.
+3.
+
+---
+
+## Pain Point Rankings (from this call)
+
+| Rank | Pain Point | Intensity (1–5) | Quote/Evidence |
+|------|-----------|-----------------|----------------|
+
+---
+
+## WTP / Value Signals
+
+| Signal | Context | Evidence Grade |
+|--------|---------|---------------|
+
+---
+
+## Key Quotes
+
+> "..."
+— {{SPEAKER}}, on {{TOPIC}}
+
+---
+
+## Hypotheses Tested
+
+| Hypothesis | Result | Evidence |
+|-----------|--------|---------|
+
+---
+
+## Learnings & Surprises
+
+<!-- What did you learn that you didn't expect? What challenges assumptions? -->
+
+---
+
+## Next Steps
+
+| Action | Owner | Due |
+|--------|-------|-----|
+
+---
+
+## Impact Assessment
+
+<!-- Does this call change any kill conditions, the hypothesis, or competitive positioning? -->
+
+- **Kill conditions affected**:
+- **Hypothesis impact**:
+- **Research gaps identified**: