Kjell Tore Guttormsen
433c2efb3d
Cold full-brief re-review (S10) reached a class the S7->S9 algorithm-stat lens never did: BLOCKER — post-feedback-monitor published as Haiku in four surfaces (README:259, skills/linkedin-studio:159 with wrong color Green too, skills/linkedin-analytics:41, agents-capability-matrix:20) while agents/post-feedback-monitor.md runs Opus. v4.0.0's Opus promotion never reached the user-facing tables. Synced all to Opus/Lime. Refreshed agents-capability-matrix.md (frozen at the v2.0 14-agent era): header 14->19, +5 missing longform agents, tier counts Opus 2->8 / Haiku 1->0, longform-gate diagram updated to the real 8-Opus-agent chain. MAJOR — de-branded docs/plan-fullspektrum-innholdsmotor.md:70 (model brand + jan-2026 asserted as fact -> no-name/no-month relevance-model phrasing). It was the only tracked survivor; the rest live in gitignored ROADMAP.md / .claude/research/ (not shipped, out of honesty scope). META — added Section 10 model-consistency guard (scripts/check-model-consistency.mjs): each agents/*.md model: must match every surface declaration AND the canonical rosters must list all 19 agents. Permanent non-vacuity self-test + e2e mutation-proven. Pre-patch sweep confirmed post-feedback-monitor was the sole drifted agent (89 model rows, 0 other mismatches). test-runner.sh 68/0/0, node --test 94/94. Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
174 lines
11 KiB
Markdown
174 lines
11 KiB
Markdown
# Agent Capability Matrix
|
||
|
||
19 specialized agents for LinkedIn content (short-form feed + long-form newsletter). Each agent has a focused responsibility, defined model, and unique color for visual identification.
|
||
|
||
## Quick Reference
|
||
|
||
| Agent | Model | Color | Primary Responsibility |
|
||
|-------|-------|-------|----------------------|
|
||
| content-optimizer | Sonnet | Blue | Optimize posts against algorithm signals |
|
||
| strategy-advisor | Sonnet | Green | Growth strategy and phase-specific guidance |
|
||
| analytics-interpreter | Sonnet | Yellow | Pattern discovery + weekly/monthly performance reports (interpret/report modes) |
|
||
| engagement-coach | Sonnet | Magenta | 5x5x5 + first-hour tactics + CEA commenting + target selection |
|
||
| content-planner | Sonnet | Cyan | Weekly/monthly content calendars |
|
||
| network-builder | Sonnet | Teal | Strategic networking and outreach |
|
||
| content-repurposer | Sonnet | Purple | Format conversion and evergreen refresh |
|
||
| trend-spotter | Sonnet | White | Trending topics and opportunity scoring |
|
||
| voice-trainer | Sonnet | Pink | Voice profile building and drift detection |
|
||
| differentiation-checker | Sonnet | Gray | Originality scoring and commodity detection |
|
||
| video-scripter | Sonnet | Violet | Video script creation with pacing and visual cues |
|
||
| post-feedback-monitor | Opus | Lime | Post-publish 48h monitoring and real-time interventions |
|
||
| fact-checker | Opus | Brown | Factual-claim verification against primary/credible sources (longform) |
|
||
| editorial-reviewer | Opus | Orange | Editor's craft gate: prosa-håndverk + narrativ-arkitektur (longform) |
|
||
| persona-reviewer | Opus | Olive | Reader-persona skeleton + resonance + hook-conversion gate (longform) |
|
||
| voice-scrubber | Opus | Red | De-AI scrub + Norwegian-chronicle voice-drift correction (longform) |
|
||
| content-reviewer | Opus | Maroon | Cold argument-integrity review (C1–C5) on a frozen draft (longform) |
|
||
| language-reviewer | Opus | Navy | Cold Norwegian-language review (L1–L5) on a frozen draft (longform) |
|
||
| fact-reviewer | Opus | Gold | Cold fact re-verification (F1–F4) + pivot-risk on a frozen draft (longform) |
|
||
|
||
## Capability Matrix
|
||
|
||
Capabilities mapped across the 14 content-production agents (the columns below). **P** = Primary, **S** = Secondary/Supporting. The five remaining agents — editorial-reviewer, voice-scrubber, content-reviewer, language-reviewer, fact-reviewer — are ordered long-form quality gates rather than content-capability agents; they are documented in **Longform Quality Gates** below and listed in the Quick Reference above.
|
||
|
||
| Capability | optimizer | strategy | analytics | engage | planner | network | repurpose | trends | voice | diff-check | video | post-monitor | fact-check | persona-rev |
|
||
|-----------|:---------:|:--------:|:---------:|:------:|:-------:|:-------:|:---------:|:------:|:-----:|:----------:|:-----:|:------------:|:----------:|:-----------:|
|
||
| Post optimization | **P** | | | | | | | | | | | | | |
|
||
| Hook analysis | **P** | | | | | | | | | | S | | | S |
|
||
| Algorithm alignment | **P** | | | S | | | | | | | S | S | | |
|
||
| Growth strategy | | **P** | | | S | | | | | | | | | |
|
||
| Phase assessment | | **P** | | | | | | | | | | | | |
|
||
| Trajectory analysis | | **P** | S | | | | | | | | | | | |
|
||
| Audience analysis | | S | **P** | | | | | | | | | | | |
|
||
| Pattern discovery | | | **P** | | | | | | | | | | | |
|
||
| Performance reports | | | **P** | | | | | | | | | | | |
|
||
| Content DNA | | | **P** | | | | | | S | | | | | |
|
||
| Engagement coaching | | | | **P** | | S | | | | | | | | |
|
||
| 5x5x5 method | | | | **P** | | S | | | | | | | | |
|
||
| Comment strategy | | | | **P** | | | | | | | | | | |
|
||
| CEA method | | | | **P** | | | | | | | | | | |
|
||
| Target identification | | | | **P** | | S | | | | | | | | |
|
||
| Content planning | | | | | **P** | | | S | | | | | | |
|
||
| Mix enforcement | | | | | **P** | | | | | | | | | |
|
||
| Gap analysis | | | | | **P** | | | | | | | | | |
|
||
| Network building | | | | S | | **P** | | | | | | | | |
|
||
| Connection scoring | | | | | | **P** | | | | | | | | |
|
||
| DM templates | | | | | | **P** | | | | | | | | |
|
||
| Format conversion | | | | | | | **P** | | | | S | | | |
|
||
| Evergreen scoring | | | | | | | **P** | | | | | | | |
|
||
| Content lifecycle | | | | | S | | **P** | | | | | | | |
|
||
| Trend scanning | | | | | S | | | **P** | | | | | | |
|
||
| First-mover assessment | | | | | | | | **P** | | | | | | |
|
||
| Angle mapping | | | | | S | | S | **P** | | | | | | |
|
||
| Voice profiling | | | | | | | | | **P** | | | | | |
|
||
| Drift detection | | | | | | | | | **P** | | | | | |
|
||
| Quarterly audit | | | | | | | | | **P** | | | | | |
|
||
| Originality scoring | | | | | | | | | | **P** | | | | |
|
||
| Commodity detection | | | | | | | | | | **P** | | | | |
|
||
| Differentiation | | | | | | | | | | **P** | | | | |
|
||
| Video scripting | | | | | | | S | | | | **P** | | | |
|
||
| Script pacing | | | | | | | | | | | **P** | | | |
|
||
| Visual cue notation | | | | | | | | | | | **P** | | | |
|
||
| Post-publish monitoring | | | | | | | | | | | | **P** | | |
|
||
| Velocity analysis | | | | | | | | | | | | **P** | | |
|
||
| Factual verification | | | | | | | | | | | | | **P** | |
|
||
| Primary-source check | | | | | | | | | | | | | **P** | |
|
||
| Persona resonance | | | | | | | | | | | | | | **P** |
|
||
| Hook-conversion gate | | | | | | | | | | | | | | **P** |
|
||
|
||
## Content Pipeline
|
||
|
||
How agents collaborate in the end-to-end content lifecycle:
|
||
|
||
```
|
||
┌─────────────┐ ┌──────────────────┐ ┌─────────────────┐
|
||
│ trend-spotter│───▸│ content-planner │───▸│ diff-checker │
|
||
│ (find topics)│ │ (plan + schedule) │ │ (originality │
|
||
└─────────────┘ └──────────────────┘ │ gate ≥51/100) │
|
||
│ └────────┬────────┘
|
||
│ │
|
||
┌──────▼──────┐ ┌───────┴────────┐
|
||
│voice-trainer│ │ FORMAT SPLIT │
|
||
│(voice check)│ └──┬──────────┬──┘
|
||
└──────┬──────┘ │ │
|
||
│ ┌───────▼───┐ ┌────▼─────────┐
|
||
│ │video- │ │content- │
|
||
└────────────▸│scripter │ │optimizer │
|
||
│(scripts) │ │(text posts) │
|
||
└───────┬───┘ └──────┬───────┘
|
||
│ │
|
||
└─────┬──────┘
|
||
┌────────────────────────────┤
|
||
│ │
|
||
┌──────▼────────────┐ ┌────────▼───────┐
|
||
│analytics- │ │ [PUBLISH] │
|
||
│interpreter │ └────────┬───────┘
|
||
│(interpret/report) │ │
|
||
└───────────────────┘ ┌────────▼───────┐
|
||
│engagement-coach│
|
||
│(5x5x5 + first │
|
||
│ hour + CEA │
|
||
│ commenting) │
|
||
└────────────────┘
|
||
```
|
||
|
||
### Longform Quality Gates (newsletter)
|
||
|
||
For longform editions, eight Opus agents run as ordered gates BEFORE lock:
|
||
|
||
```
|
||
draft
|
||
─▸ fact-checker (primary-source verification, post-cutoff web search)
|
||
─▸ editorial-reviewer (craft: prosa-håndverk + narrativ-arkitektur, Step 5.5)
|
||
─▸ persona-reviewer (skeleton → resonance → hook-conversion)
|
||
─▸ voice-scrubber (de-AI + Norwegian-chronicle voice)
|
||
─▸ headless review (Step 6.5 — COLD on the frozen draft):
|
||
content-reviewer (argument integrity C1–C5)
|
||
language-reviewer (Norwegian language L1–L5)
|
||
fact-reviewer (cold re-verification F1–F4 + pivot-risk)
|
||
─▸ LOCK ─▸ delivery
|
||
```
|
||
|
||
### Parallel Support Agents
|
||
|
||
These agents operate independently and feed into the pipeline at multiple points:
|
||
|
||
```
|
||
strategy-advisor ──────▸ Macro-level planning and phase guidance
|
||
analytics-interpreter ─▸ Pattern discovery + periodic reports feeding back into planning
|
||
network-builder ───────▸ Relationship building amplifying content reach
|
||
content-repurposer ────▸ Post-publish: extends content lifecycle
|
||
```
|
||
|
||
## Which Agent Do I Need?
|
||
|
||
| Scenario | Agent | Command |
|
||
|----------|-------|---------|
|
||
| "I want to write a post" | content-optimizer | `/linkedin:post` |
|
||
| "What should I post about?" | content-planner, trend-spotter | `/linkedin:pipeline` |
|
||
| "Make this post better" | content-optimizer | `/linkedin:post` |
|
||
| "Is this original enough?" | differentiation-checker | `/linkedin:pipeline` |
|
||
| "Plan my week's content" | content-planner | `/linkedin:batch` |
|
||
| "Am I on track this week?" | — | `/linkedin:calendar` |
|
||
| "How did I do this week?" | analytics-interpreter (report mode) | `/linkedin:report` |
|
||
| "Analyze my LinkedIn data" | analytics-interpreter (interpret mode) | `/linkedin:analyze` |
|
||
| "What's my LinkedIn strategy?" | strategy-advisor | `/linkedin:strategy` |
|
||
| "Help me engage more" | engagement-coach | `/linkedin:strategy` |
|
||
| "Who should I comment on?" | engagement-coach | `/linkedin:strategy` |
|
||
| "Build my network" | network-builder | `/linkedin:strategy` |
|
||
| "Does this sound like me?" | voice-trainer | `/linkedin:post` |
|
||
| "Repurpose my best post" | content-repurposer | `/linkedin:pipeline` |
|
||
| "What's trending in my field?" | trend-spotter | `/linkedin:pipeline` |
|
||
| "Audit my content strategy" | analytics-interpreter, strategy-advisor | `/linkedin:audit` |
|
||
| "How do I monetize?" | strategy-advisor | `/linkedin:monetize` |
|
||
| "Create a video script" | video-scripter | `/linkedin:video` |
|
||
| "Turn this post into a video" | video-scripter, content-repurposer | `/linkedin:video` |
|
||
| "Script a talking head video" | video-scripter | `/linkedin:video` |
|
||
| "Verify facts in this draft" | fact-checker | `/linkedin:newsletter` (longform) |
|
||
| "Will this land with my readers?" | persona-reviewer | `/linkedin:newsletter` (longform) |
|
||
|
||
## Model Selection Rationale
|
||
|
||
| Model | Agents | Why |
|
||
|-------|--------|-----|
|
||
| **Opus** | 8 agents (fact-checker, editorial-reviewer, persona-reviewer, voice-scrubber, content-reviewer, language-reviewer, fact-reviewer, post-feedback-monitor) | Longform judgment + 48h post-monitoring: factual verification, craft, resonance, voice, cold adversarial re-review, real-time intervention |
|
||
| **Sonnet** | 11 agents | Complex reasoning: optimization, strategy, analysis, scoring, scripting, comment targeting |
|