@geraldmaron/construct 1.0.16 → 1.0.17

package/skills/strategy/competitive-landscape.md

@@ -1,6 +1,8 @@
 ---
 name: strategy-competitive-landscape
 description: Use when the team needs a structured read on market positioning before committing direction.
+inputs: [competitor-data, market-signals]
+artifactType: competitive-analysis
 ---
 # Competitive Landscape Analysis
 

package/skills/strategy/jobs-to-be-done.md

@@ -0,0 +1,38 @@
+---
+name: strategy-jobs-to-be-done
+description: "Use when: user research needs to uncover the job a user is hiring a product to do — the underlying progress they seek — rather than feature requests or stated preferences."
+inputs: [user-signals-or-interviews]
+artifactType: jtbd-summary
+---
+# Jobs To Be Done
+
+Use this skill when user research needs to surface the underlying job a user is trying to get done, not the features they ask for.
+
+Users describe solutions ("add a dark mode"); the job is the progress they're trying to make and the context that triggers it. Building to stated features without the job produces local optima nobody hires.
+
+## Steps
+
+1. **Anchor on a real episode**, not a hypothetical. "Tell me about the last time you…" — recent, specific, remembered.
+2. **Trace the timeline**: the trigger (what changed that made them act), the struggle with the current approach, what they tried, and the outcome they were after.
+3. **Separate the job from the solution.** "I want X feature" → ask "what would X let you do that you can't now?" Keep climbing until you reach a stable goal that doesn't reference your product.
+4. **Capture the forces**: push (problems with the status quo), pull (attraction of a new way), and anxieties/habits that hold them back. Adoption is decided by these four forces.
+5. **Watch for non-consumption**: people who have the job but use nothing, or a manual workaround — often the clearest signal of the job.
+6. **State the job** as `When <situation>, I want to <motivation>, so I can <expected outcome>` — sourced to the episodes, not invented.
+
+## Output shape
+
+```
+## Job statement(s)
+- When <situation>, I want to <motivation>, so I can <outcome> — evidence: <episode refs>
+
+## Forces
+- push: … · pull: … · anxiety: … · habit: …
+
+## Non-consumption / workarounds observed
+```
+
+## Verification bar
+
+- Each job statement traces to a specific recounted episode, not a generalization or an assumption about "users".
+- Solution language ("add X") is climbed back to the underlying job, not recorded as the job.
+- Forces and non-consumption come from what was observed/said, labeled where inferred.

package/skills/strategy/market-research-methods.md

@@ -1,6 +1,8 @@
 ---
 name: strategy-market-research-methods
 description: Use when the team needs to validate assumptions before committing resources, or when a decision is being made on vibes rather than signal.
+inputs: [market-signals, user-signals]
+artifactType: market-brief
 ---
 # Market Research Methods
 

package/skills/strategy/narrative-arc.md

@@ -1,6 +1,8 @@
 ---
 name: strategy-narrative-arc
 description: Use when the argument must move people, not just inform them. Strategic narrative is the difference between a document that gets read and one that changes what people do.
+inputs: [market-signals, user-signals]
+artifactType: positioning
 ---
 # Strategic Narrative Construction
 

package/skills/strategy/pricing-positioning.md

@@ -1,6 +1,8 @@
 ---
 name: strategy-pricing-positioning
 description: Use when the team needs to set price, adjust positioning, or defend either to stakeholders.
+inputs: [market-signals, competitor-data]
+artifactType: positioning
 ---
 # Pricing and Positioning
 

package/skills/utility/clean-code.md

@@ -1,6 +1,8 @@
 ---
 name: utility-clean-code
 description: Patterns and heuristics for identifying and removing AI-generated code smells. Use when cleaning up output from AI coding tools.
+inputs: [code]
+artifactType: refactor
 ---
 # AI Slop Removal
 

package/specialists/prompts/cx-engineer.md

@@ -44,7 +44,7 @@ Construct may dispatch you in response to a `handoff.received`, `incident.handof
 **Fence (declared in specialists/role-manifests.json → engineer):**
 - Allowed paths: `lib/**`, `bin/**`, `src/**`, `app/**`, `tests/**`, `docs/**`
 - Allowed bd labels: `bug`, `feature`, `task`, `engineering`, `fix`
-- Approval required: any commit, any push, any edit to protected files (`specialists/registry.json`, `install.sh`, `claude/settings.template.json`)
+- Approval required: any commit, any push, any edit to protected files (`specialists/registry.json`, `lib/setup.mjs`, `claude/settings.template.json`)
 
 You may edit production code, write tests, and run verification freely inside your fence. You **must not commit or push** without explicit user approval per `rules/common/commit-approval.md`. Stage edits, run verification, and stop.
 

package/specialists/registry.json

@@ -50,7 +50,8 @@
       "skills": [
         "ai/orchestration-workflow",
         "operating/orchestration-reference",
-        "operating/change-management"
+        "operating/change-management",
+        "operating/unstructured-triage"
       ],
       "perspective": {
         "bias": "Over-routing to cx-engineer, false simplicity, plans where every task runs in parallel",
@@ -200,7 +201,8 @@
         ]
       },
       "skills": [
-        "docs/research-workflow"
+        "docs/research-workflow",
+        "strategy/jobs-to-be-done"
       ],
       "perspective": {
         "bias": "Ideal-user personas, invisible friction, 'users will figure it out'",
@@ -291,7 +293,8 @@
       },
       "skills": [
         "docs/research-workflow",
-        "devops/dependency-management"
+        "devops/dependency-management",
+        "docs/transcript-synthesis"
       ],
       "perspective": {
         "bias": "Undated sources, version-specific claims without citation, blog posts treated as authoritative",
@@ -389,7 +392,8 @@
       },
       "skills": [
         "devops/observability",
-        "devops/database"
+        "devops/database",
+        "operating/raw-data-structuring"
       ],
       "perspective": {
         "bias": "Vanity metrics, post-hoc success definitions, averages that hide distributions",
@@ -645,7 +649,8 @@
         ]
       },
       "skills": [
-        "quality-gates/review-work"
+        "quality-gates/review-work",
+        "quality-gates/premortem"
       ],
       "perspective": {
         "bias": "Plans that are too elegant, 'unlikely' failure modes, scope drift with stable acceptance criteria",
@@ -1094,7 +1099,8 @@
       "skills": [
         "docs/adr-workflow",
         "docs/runbook-workflow",
-        "docs/init-docs"
+        "docs/init-docs",
+        "docs/memo-and-decision-capture"
       ],
       "perspective": {
         "bias": "Completed work with no context update, decisions 'everyone understands' but nobody wrote down",
@@ -1193,7 +1199,8 @@
         ]
       },
       "skills": [
-        "frontend-design/accessibility"
+        "frontend-design/accessibility",
+        "frontend-design/screen-reader-testing"
       ],
       "perspective": {
         "bias": "ARIA added to pass audits, keyboard navigation never tested, contrast that 'looks fine' but fails the spec",
@@ -1280,7 +1287,8 @@
         ]
       },
       "skills": [
-        "ai/prompt-optimizer"
+        "ai/prompt-optimizer",
+        "ai/trace-triage"
       ],
       "perspective": {
         "bias": "Stable-looking medians hiding variance, trace analyses looking only at successes, promotions on too few traces",
@@ -1322,7 +1330,8 @@
       "skills": [
         "devops/data-engineering",
         "devops/database",
-        "devops/observability"
+        "devops/observability",
+        "operating/raw-data-structuring"
       ],
       "perspective": {
         "bias": "Non-idempotent pipelines, unwritten data contracts, quality gates added after the first corruption incident",

package/specialists/role-manifests.json

@@ -73,7 +73,7 @@
         "allowedPaths": ["lib/**", "bin/**", "src/**", "app/**", "tests/**", "docs/**"],
         "allowedBdLabels": ["bug", "feature", "task", "engineering", "fix"],
         "allowedCommands": ["bd create", "bd note", "bd update", "bd link", "bd label", "bd close", "node", "npm", "git status", "git diff", "git log"],
-        "approvalRequired": ["commit", "push", "edit:specialists/registry.json", "edit:install.sh", "edit:claude/settings.template.json"]
+        "approvalRequired": ["commit", "push", "edit:specialists/registry.json", "edit:lib/setup.mjs", "edit:claude/settings.template.json"]
       },
       "outputs": {
         "bdLabels": ["engineering", "fix"],