refactor(linkedin-studio): wire or delete 11 orphan agents (case-by-case) — 9 here, 2 in Steps 14/16

Resolves the orphan-agent audit finding by the locked default: wire all, no deletions, so the agent count stays 19. Per agent, added Task to the target command's allowed-tools and a coherent 'subagent_type: linkedin-studio:<name>' delegation at a real point in the command's flow (not a token grep-match).

Wired (agents 1-9 of 11): video-scripter -> video.md (Step 4); content-optimizer -> post.md (Step 7 refinement) + ab-test.md (2a.4 optimized challenger); analytics-interpreter -> report.md (Step 7, report mode) + analyze.md (Step 2, interpret mode); content-planner -> batch.md (Step 2) + pipeline.md (Step 1); trend-spotter -> batch.md (Step 1) + pipeline.md (Step 1); network-builder -> outreach.md (Step 3a); strategy-advisor -> strategy.md (Step 3); voice-trainer -> setup.md (Step 3a); post-feedback-monitor -> calendar.md (publish action, 48h monitor).

Deferred to their dedicated steps: #10 differentiation-checker -> Step 14 (short-form de-AI gate), #11 engagement-coach -> Step 16 (first-hour command). Namespaced subagent_type form requires a session reload before the wired agents are invokable.

Verify: each of the 9 has >=1 invocation in commands/; structural lint 61/61 (counts 19/26/25/6 intact); agent-fixtures 35/35; hook tests 62/62. Three-doc + version reconciliation deferred to Step 21 per the locked plan [skip-docs].

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

plugins/linkedin-studio/commands/ab-test.md

@@ -15,6 +15,7 @@ allowed-tools:
   - Write
   - Bash
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn A/B Testing Command
@@ -115,6 +116,8 @@ Document the test variant:
 - Ensure ONLY the target variable changes
 - Be concrete: "Provocative question hooks, e.g., 'What if AI readiness has nothing to do with technology?'"
 
+When variant B is an *optimized* challenger to the control (rather than a simple variable swap), delegate the rewrite to the `content-optimizer` agent — invoke it via `Task` with `subagent_type: linkedin-studio:content-optimizer` (foreground, from this command layer), holding every dimension constant except the target variable so the test stays clean.
+
 ### 2a.5: Plan Execution
 
 Generate a posting schedule that alternates A/B on comparable days:

plugins/linkedin-studio/commands/analyze.md

@@ -9,6 +9,7 @@ description: |
 allowed-tools:
   - Read
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Performance Analysis & Troubleshooting
@@ -37,6 +38,8 @@ Use AskUserQuestion to understand the situation:
 
 ## Step 2: Gather Data
 
+If imported analytics data exists (`assets/analytics/`), delegate audience-pattern discovery to the `analytics-interpreter` agent (interpret mode) — invoke it via `Task` with `subagent_type: linkedin-studio:analytics-interpreter` (foreground, from this command layer) — to ground the diagnosis in what the data actually shows before relying on self-report.
+
 Based on their answer, ask relevant follow-up questions:
 
 ### If Reach Dropped Suddenly

plugins/linkedin-studio/commands/batch.md

@@ -40,10 +40,12 @@ Use AskUserQuestion:
 3. **Use existing plan** — Follow the weekly plan already created
 4. **Mix it up** — Diverse topics across pillars
 
-If they choose a theme, help them identify 3-5 unique angles from `references/thought-leadership-angles.md`.
+If they choose a theme, help them identify 3-5 unique angles from `references/thought-leadership-angles.md`. For timely angles, delegate to the `trend-spotter` agent — invoke it via `Task` with `subagent_type: linkedin-studio:trend-spotter` (foreground, from this command layer) — to surface trending topics and score their relevance against the user's pillars.
 
 ## Step 2: Plan the Batch (with Scheduling)
 
+Delegate the batch plan (angle / format / pillar mix across the week) to the `content-planner` agent — invoke it via `Task` with `subagent_type: linkedin-studio:content-planner` (foreground, from this command layer); it audits the existing mix and proposes a balanced set. This command owns scheduling the result.
+
 Read `${CLAUDE_PLUGIN_ROOT}/references/scheduling-strategy.md` for optimal posting slots.
 
 Calculate scheduled dates based on `weekly_goal` from state:

plugins/linkedin-studio/commands/calendar.md

@@ -13,6 +13,7 @@ allowed-tools:
   - Write
   - Edit
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Content Calendar
@@ -148,7 +149,7 @@ First Hour:
 
 48-Hour Check-In:
 - Run /linkedin:analyze after 48 hours to review performance
-- Or use post-feedback-monitor agent for real-time tracking
+- For real-time tracking, delegate to the `post-feedback-monitor` agent — invoke it via `Task` with `subagent_type: linkedin-studio:post-feedback-monitor` (foreground, from this command layer) to watch the post's first-48h engagement and flag anomalies early
 ```
 
 **3f. Ask about more.** Use AskUserQuestion:

plugins/linkedin-studio/commands/outreach.md

@@ -19,6 +19,7 @@ allowed-tools:
   - WebFetch
   - AskUserQuestion
   - Write
+  - Task
 ---
 
 # LinkedIn Outreach Orchestrator (Collaborations + Speaking)
@@ -168,6 +169,8 @@ Ask the user (via AskUserQuestion):
 3. Someone recently engaged with my content
 4. I want to build a collaboration network from scratch
 
+For options 2 and 4 (finding partners or building a network from scratch), delegate discovery and prioritization to the `network-builder` agent — invoke it via `Task` with `subagent_type: linkedin-studio:network-builder` (foreground, from this command layer); it identifies niche-relevant connections and applies the scoring criteria below.
+
 #### Partner Scoring Criteria
 
 For each potential partner, evaluate:

plugins/linkedin-studio/commands/pipeline.md

@@ -46,6 +46,8 @@ Otherwise, check state file for `next_planned_topic`:
   1. I have an idea already
   2. Generate ideas for me
 
+To situate the post in the broader plan — does it fill a content-mix gap or repeat a recent pillar? — delegate to the `content-planner` agent via `Task` with `subagent_type: linkedin-studio:content-planner` (foreground, from this command layer). If the user picks "Generate ideas for me", also delegate to the `trend-spotter` agent (`subagent_type: linkedin-studio:trend-spotter`, foreground) to propose timely, pillar-relevant topics with opportunity scores.
+
 ## Step 2: Draft
 
 Once topic is chosen, create the draft:

plugins/linkedin-studio/commands/post.md

@@ -13,6 +13,7 @@ allowed-tools:
   - WebFetch
   - Bash
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Post Creation Workflow
@@ -156,6 +157,8 @@ Do NOT use AskUserQuestion here. Simply state:
 
 Wait for the user to respond naturally. Iterate until they're satisfied or they indicate the post is ready.
 
+When a refinement calls for systematic optimization — hook strength, structure, or engagement mechanics rather than a one-line tweak — delegate to the `content-optimizer` agent: invoke it via `Task` with `subagent_type: linkedin-studio:content-optimizer` (foreground, from this command layer), then apply its suggestions to the draft. For light touch-ups, edit inline.
+
 ## Step 8: Pre-Publish Reminder
 
 Before they post, remind them:

plugins/linkedin-studio/commands/report.md

@@ -11,6 +11,7 @@ allowed-tools:
   - Read
   - Glob
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Analytics Weekly Report
@@ -303,7 +304,7 @@ Example recommendations:
 
 ## Step 7: Generate Actionable Recommendations
 
-Based on the report data, provide 2-3 specific, actionable recommendations:
+Delegate interpretation to the `analytics-interpreter` agent (report mode) — invoke it via `Task` with `subagent_type: linkedin-studio:analytics-interpreter` (foreground, from this command layer), passing the generated report data; it surfaces the patterns behind the numbers. Based on the report data and the agent's reading, provide 2-3 specific, actionable recommendations:
 
 **Framework for recommendations:**
 

plugins/linkedin-studio/commands/setup.md

@@ -13,6 +13,7 @@ allowed-tools:
   - Glob
   - Write
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Plugin Setup & Personalization
@@ -82,6 +83,8 @@ Based on their answer, run the corresponding sub-workflow below.
 
 **Goal:** Populate `assets/voice-samples/authentic-voice-samples.md` with real voice data.
 
+**Delegate the analysis + profile construction to the `voice-trainer` agent** — invoke it via `Task` with `subagent_type: linkedin-studio:voice-trainer` (foreground, from this command layer). The agent performs the pattern detection and extraction (steps 2–3 below) and returns the structured voice profile; this command owns collecting the samples (step 1) and writing the profile back to disk (steps 4–6).
+
 1. Ask the user to paste 3-5 of their best LinkedIn posts (or any professional writing samples)
 2. Analyze the samples for:
    - **Sentence structure:** Short/long, simple/complex, varied?

plugins/linkedin-studio/commands/strategy.md

@@ -15,6 +15,7 @@ allowed-tools:
   - Grep
   - WebFetch
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Growth Strategy
@@ -150,7 +151,7 @@ Based on follower count, place user in the right phase:
 
 ## Step 3: Provide Phase-Specific Strategy
 
-Based on their phase, provide detailed guidance.
+Delegate the growth recommendation to the `strategy-advisor` agent — invoke it via `Task` with `subagent_type: linkedin-studio:strategy-advisor` (foreground, from this command layer), passing the detected phase and trajectory; it matches the user to the right roadmap phase and prioritizes high-impact actions. Use its output to provide detailed guidance.
 
 ### For Phase 0-1 (Under 3,000)
 

plugins/linkedin-studio/commands/video.md

@@ -12,6 +12,7 @@ allowed-tools:
   - Write
   - Bash
   - AskUserQuestion
+  - Task
 ---
 
 # LinkedIn Video Script Creation Workflow
@@ -77,7 +78,7 @@ Follow the same flow as `/linkedin:post`:
 
 ## Step 4: Generate Script
 
-Delegate to the `video-scripter` agent for script generation. The agent will:
+Delegate script generation to the `video-scripter` agent — invoke it via `Task` with `subagent_type: linkedin-studio:video-scripter` (foreground, from this command layer). The agent will:
 
 1. Calculate word budget based on selected length (duration × 2.5 wps)
 2. Select the appropriate script template from `references/video-strategy-guide.md`