# Autobrand Skills

\---

name: autobrand\
description: Autonomous brand optimization loop. Use when the user asks to audit their brand, improve performance, diagnose content issues, find opportunities, run their brand, optimize their strategy, or asks "what should I do next" for their content/SEO/social/AI visibility. This skill provides the decision framework - the brain behind the tools.

***

## AutoBrand: Self-Improving Brand Loop

This skill teaches you HOW to interpret brand performance data and WHAT actions to take. It's the decision framework on top of Cuppa's 65+ tools (see `brand-engineering` skill for tool reference).

Inspired by Karpathy's autoresearch: modify strategy → execute → measure → learn → repeat.

### The Loop

```
1. MEASURE   → Pull performance data
2. DIAGNOSE  → Compare against benchmarks, find problems
3. PRESCRIBE → Choose specific actions for each problem
4. EXECUTE   → Run the tools
5. REPEAT    → Wait for signals, measure again
```

### Step 1: Measure

Always start by pulling the full picture:

```
get_brand (site_id)             → Brand DNA, competitors, industry
get_brand_performance (site_id) → KPIs + Visibility Score + Quick Wins
get_content_health (site_id)    → Top pages, decay, cannibalization
get_ai_visibility (site_id)     → LLM citations, gaps
get_backlinks (site_id)         → DR, referring domains
get_social_analytics (site_id)  → Live social analytics: impressions, engagements, reach, followers, best times, content decay, top posts
```

### Step 2: Diagnose

Use the brand context (industry, stage, competitors) to interpret the numbers. The Brand DNA tells you WHO the brand is. The benchmarks below tell you WHERE they should be.

#### Organic Search Benchmarks

| Metric                     | Poor   | Okay   | Good   | Great                  |
| -------------------------- | ------ | ------ | ------ | ---------------------- |
| Avg CTR (all positions)    | < 1.5% | 1.5-3% | 3-5%   | > 5%                   |
| CTR for position 1-3       | < 5%   | 5-15%  | 15-30% | > 30%                  |
| Striking distance keywords | 0-5    | 5-15   | 15-50  | 50+ (huge opportunity) |
| Content grade avg          | < 50   | 50-65  | 65-80  | 80+                    |
| Pages with decay (14d)     | > 30%  | 15-30% | 5-15%  | < 5%                   |

#### Domain Authority Benchmarks (by stage)

| Brand Stage            | Expected DR | Action If Below                               |
| ---------------------- | ----------- | --------------------------------------------- |
| New (< 6 months)       | 5-15        | Normal. Focus on content first.               |
| Early (6-18 months)    | 15-30       | Build links. Guest posts + marketplace.       |
| Growing (18-36 months) | 30-50       | Accelerate links. Expert reviews for E-E-A-T. |
| Established (3+ years) | 50-70       | Maintain. Focus on content quality.           |
| Market leader          | 70+         | Defend. Monitor competitors.                  |

#### AI Visibility Benchmarks

| Metric                          | Invisible | Emerging | Visible | Strong |
| ------------------------------- | --------- | -------- | ------- | ------ |
| Total LLM citations             | 0-2       | 3-10     | 10-50   | 50+    |
| Citation gaps vs top competitor | > 20      | 10-20    | 5-10    | < 5    |
| Unique prompts you appear in    | 0-5       | 5-20     | 20-50   | 50+    |

#### Social Benchmarks

| Metric               | Low      | Average  | Good     | Excellent |
| -------------------- | -------- | -------- | -------- | --------- |
| Avg engagement rate  | < 0.5%   | 0.5-2%   | 2-5%     | > 5%      |
| Impressions per post | < 500    | 500-2K   | 2K-10K   | 10K+      |
| Consistent posting   | < 2/week | 2-3/week | 4-5/week | Daily     |

#### Brand Visibility Score

| Score  | Meaning                             | Priority                           |
| ------ | ----------------------------------- | ---------------------------------- |
| 0-20   | Invisible. Start from scratch.      | Content volume + brand DNA         |
| 20-40  | Early stage. Foundations need work. | Fix content quality + start links  |
| 40-60  | Growing. Visible but gaps remain.   | Optimize existing + fill gaps      |
| 60-80  | Strong. Fine-tuning needed.         | Defend positions + expand channels |
| 80-100 | Dominant. Maintain and defend.      | Monitor competitors + experiment   |

### Step 3: Prescribe

Match each diagnosis to specific actions. Work top to bottom (highest impact first).

#### Priority 1: Fix What's Broken

| Signal                           | Diagnosis                | Action                                                                  |
| -------------------------------- | ------------------------ | ----------------------------------------------------------------------- |
| Pages losing > 20% traffic (14d) | Content decay            | `optimize_content` on each decaying page, then `publish_content`        |
| Keyword cannibalization detected | Multiple pages competing | Consolidate: pick winner, redirect or merge losers, `add_internal_link` |
| Content grade < 60 on key pages  | Below SERP quality       | `grade_content` → `optimize_content` → `grade_content` loop until 70+   |
| Broken CMS publish               | Integration issue        | Check integration settings, re-publish                                  |

#### Priority 2: Capture Quick Wins

| Signal                                                      | Diagnosis                 | Action                                                        |
| ----------------------------------------------------------- | ------------------------- | ------------------------------------------------------------- |
| Striking distance keywords (pos 4-20) with existing content | Almost ranking            | `optimize_content` on those articles, focus on target keyword |
| High impressions / low CTR (pos 1-5)                        | Bad title or meta         | `update_article` with better title/meta description           |
| Striking distance with NO content                           | Missing content           | `create_content` targeting that keyword                       |
| High-performing social format                               | Winning format identified | Double down: more of that format (carousel, video, hook type) |

#### Priority 3: Build Authority

| Signal                   | Diagnosis            | Action                                                           |
| ------------------------ | -------------------- | ---------------------------------------------------------------- |
| DR below stage benchmark | Weak link profile    | Link Marketplace: target DR 40+ sites, dofollow, your niche      |
| Low E-E-A-T signals      | No expert validation | Expert Reviews for pillar articles                               |
| Few referring domains    | Narrow link profile  | Diversify: different link sources, categories                    |
| No AI citations          | LLMs don't know you  | `create_knowledge` with structured FAQ, comparison, product data |

#### Priority 4: Expand Reach

| Signal                       | Diagnosis                            | Action                                                            |
| ---------------------------- | ------------------------------------ | ----------------------------------------------------------------- |
| Citation gaps vs competitors | Competitors cited, you're not        | Create content targeting those exact prompts/topics               |
| Social reach plateaued       | Audience capped on current platforms | Test new platform: `generate_social` on untried platform          |
| No video content             | Missing video channel                | `generate_video_script` + `create_video` for top topics           |
| No newsletter                | Email channel unused                 | `generate_newsletter` + `update_content_strategy` (enable weekly) |
| Content gaps from GSC        | Audience searching, you're not there | `create_content` for each gap topic                               |

#### Priority 5: Optimize and Compound

| Signal                             | Diagnosis            | Action                                                               |
| ---------------------------------- | -------------------- | -------------------------------------------------------------------- |
| Top-performing articles            | Winners identified   | `generate_social_multi` to amplify, `generate_carousel` for LinkedIn |
| High-grade articles without images | Missing visual layer | `generate_images_bulk` with brand colors                             |
| Published articles without social  | Under-distributed    | `generate_social_multi` for each, `schedule_social` across week      |
| Articles without internal links    | Orphan content       | `add_internal_link` to connect to pillar pages                       |

### Step 4: Execute

Run the prescribed actions using the tools. See `brand-engineering` skill for full tool reference and syntax.

Key execution patterns:

**Content quality loop:**

```
grade_content → check score
  If < 70: optimize_content → grade_content again
  If >= 70: proceed to publish
```

**Full content pipeline:**

```
create_content → poll get_content_status → grade_content
→ optimize_content (if needed) → generate_images_bulk
→ publish_content → generate_social_multi → schedule_social
```

**Authority building:**

```
get_backlinks → check DR vs benchmark
→ Link Marketplace (UI) for guest post opportunities
→ Expert Reviews (UI) for pillar articles
→ create_knowledge for AI visibility
→ refresh_link_sitemaps after publishing
```

### Step 5: Repeat

#### Feedback timing

| Signal Type       | Available After | What To Check                         |
| ----------------- | --------------- | ------------------------------------- |
| Content grade     | Immediate       | Did optimize\_content improve score?  |
| Social engagement | 24-48 hours     | Which platforms/formats got traction? |
| GA4 sessions      | 24-48 hours     | Is new content getting traffic?       |
| GSC rankings      | 3-14 days       | Are positions improving?              |
| AI citations      | 7-30 days       | Are LLMs starting to cite you?        |
| DR changes        | 30-60 days      | Is link building working?             |

#### What to check each cycle

```
get_brand_performance → Is Visibility Score trending up?
get_content_health → Are fewer pages decaying?
get_ai_visibility → Are citation gaps closing?
get_social_analytics → Is engagement rate improving?
get_backlinks → Is DR growing?
```

#### When to change strategy

* **3 cycles with no Visibility Score improvement** → Re-examine priorities. You may be optimizing the wrong channel.
* **Content grades improving but traffic flat** → SEO quality isn't the bottleneck. Focus on links or distribution.
* **Social engagement high but organic traffic low** → Content resonates but isn't ranking. Focus on SEO optimization.
* **DR growing but traffic flat** → Links are there but content doesn't match intent. Refresh content.
* **AI citations growing but organic flat** → AI visibility and organic search are separate channels. Both are working, continue.

### Brand Type Modifiers

Adjust priorities based on brand type:

**SaaS / Software:**

* Highest priority: AI visibility (your buyers ask ChatGPT)
* Key content: comparison pages, alternatives pages, how-to guides
* Social focus: LinkedIn, X (Twitter)
* DR target: 40+ within 18 months

**E-commerce / DTC:**

* Highest priority: Organic traffic (direct purchase intent)
* Key content: product guides, buying guides, reviews
* Social focus: Instagram, Pinterest, TikTok
* DR target: 30+ within 12 months

**Agency / Services:**

* Highest priority: E-E-A-T (trust signals)
* Key content: case studies, thought leadership, methodology
* Social focus: LinkedIn
* DR target: 50+ (credibility critical)

**Local Business:**

* Highest priority: Local SEO + Google Business
* Key content: location pages, service pages, local guides
* Social focus: Facebook, Instagram
* DR target: 20+ (local competition is lower)

**Media / Publisher:**

* Highest priority: Content volume + freshness
* Key content: news, analysis, trending topics
* Social focus: X (Twitter), Threads
* DR target: 60+ (competing with media sites)

### Example Diagnosis

Agent receives this data for a SaaS brand (18 months old):

```
Visibility Score: 34
Organic clicks: 890/month (↓8%)
CTR: 1.8%
DR: 22
AI citations: 2
Social reach: 3,200/month
Striking distance keywords: 31
Content decay: 12 pages losing > 20%
Citation gaps: 18 prompts where competitors cited, you're not
Avg content grade: 62
```

Agent should reason:

1. **Visibility Score 34** → "Early stage, foundations need work" (benchmark: 20-40)
2. **DR 22 at 18 months** → Below benchmark (should be 30+). Link building is critical.
3. **CTR 1.8%** → Below average (benchmark: 1.5-3% is "okay"). Titles/metas need work.
4. **AI citations: 2** → Invisible to LLMs. Need knowledge base content.
5. **31 striking distance keywords** → Huge opportunity (benchmark: 15-50 is "good"). Prioritize optimizing existing content over creating new.
6. **12 decaying pages** → Fix these FIRST before creating new content.
7. **Avg grade 62** → Below "good" (65-80). Content quality is a bottleneck.
8. **18 citation gaps** → Competitors are visible in AI, you're not. Create structured content.

Prescribed action plan:

```
IMMEDIATE (this week):
1. optimize_content on 12 decaying pages → grade until 70+
2. optimize_content on top 10 striking distance articles

NEXT WEEK:
3. create_content for 5 striking distance keywords with no article
4. create_knowledge with FAQ + comparison data (for AI citations)
5. generate_social_multi for all refreshed articles

ONGOING:
6. Link Marketplace: target 3-5 guest posts/month (DR 40+ sites)
7. Expert Reviews: submit 2 pillar articles for review
8. generate_newsletter weekly from published content
```


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://learn.cuppa.ai/agent-marketing/autobrand-skills.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.