Content Miner

Content Miner

Mine the 876-[[episodes|episode archive]] (793 MFM + 83 MoneyWise) for content opportunities backed by archive evidence. The user provides context — trending topics, audience questions, content gaps — and gets ranked ideas with specific episode references, quotes, and angles.

When to Use

The user wants content ideas grounded in their existing archive. They might say:

• "What should I talk about next?"
• "What's in my archive about [topic]?"
• "Which guests should I bring back?"
• "This topic is trending — do I have anything on it?"
• "What topics have I mentioned but never really covered?"
• "Find me revisit-worthy takes from old episodes"

How It Works

This skill has four mining modes. Ask the user which mode they want, or suggest one based on their question. If they say something general like "find me content ideas," run Gap Mining first, then offer the other modes.

No external trend awareness. The user provides the trend context. This skill's job is to search the archive, not the internet.

Mode 1: Gap Mining

Purpose: Find topics mentioned across episodes but never deep-dived.

Trigger phrases: "What haven't I covered?" / "What topics am I missing?" / "Find gaps in my content"

Workflow

1. Read references/indexes/by-topic.md — this maps topics to episode IDs
2. Read references/indexes/by-framework.md — this lists the 47 frameworks with dedicated articles
3. Cross-reference: which topics appear in 3+ episodes in the topic index but have no corresponding framework article in the framework index?
4. For each gap found, grep references/transcripts/ for the topic keyword to confirm it appears in multiple transcripts and isn't just an index artifact
5. Read 2-3 relevant transcript sections to find the best quotes and angles

Output: Gap Analysis Table

For each gap (aim for 5-10):

Include a representative quote from the strongest episode for each gap.

Mode 2: Revisit Mining

Purpose: Find takes and predictions from 2+ years ago worth revisiting now.

Trigger phrases: "Was I right about X?" / "What did I say about [topic] years ago?" / "Find revisit-worthy takes"

Workflow

1. User provides a topic (or ask them for one — e.g., "AI," "crypto," "remote work," "newsletters")
2. Grep references/transcripts/ for the topic across the archive
3. Focus on older episodes — look at episode filenames/dates to identify content from 2+ years ago
4. Read the relevant transcript sections for strong, specific takes or predictions (not vague observations)
5. For each revisit candidate, note: what was the original take? What has changed since? What's the revisit angle?

Output: Revisit Candidates

For each candidate (aim for 3-5):

Episode: [ID / title] Original Take: "[Direct quote from transcript]" What's Changed: [Brief description of what happened since — the user can fill in details since they know the current landscape better than the archive does] Revisit Angle: [Specific framing for a new episode — "I was right about X," "I was wrong about Y," "This aged surprisingly well," "Here's what nobody saw coming"]

Mode 3: Callback Mining

Purpose: Identify guests worth bringing back based on archive presence and time elapsed.

Trigger phrases: "Which guests should I bring back?" / "Who haven't I had on in a while?" / "Guest ideas"

Workflow

1. Read references/indexes/by-guest.md — this lists guests with their episodes, frameworks, and key topics
2. Read references/people/ profiles for guests with high archive presence (multiple episodes, associated frameworks, or skill references)
3. Identify guests who:
   • Appeared in 2+ episodes OR have a dedicated framework/skill
   • Haven't appeared recently (check episode IDs — older-format filenames like title_slug tend to be older episodes, YouTube IDs like k7UycsE_QoA tend to be newer)
   • Run businesses or projects that have likely evolved significantly
4. For each callback candidate, read their people profile and note what they were known for, then identify what might have changed

Output: Callback Candidates

For each candidate (aim for 5-8):

Guest: [Name] Last Appearance: [Episode ID / approximate date] Known For: [Their business/framework from the archive] What's Likely Changed: [Business growth, pivot, new venture, market shift — based on what the archive says plus reasonable inference] Callback Topic: "[Specific episode premise — e.g., 'Have [[frameworks/alex-hormozi|Alex Hormozi]] back to talk about what [[frameworks/acquisition-com|Acquisition.com]] looks like at $500M+ portfolio']"

Mode 4: Clip Mining

Purpose: User has a trending topic and wants to know what's already in the archive about it.

Trigger phrases: "This is trending — what do I have on it?" / "Find clips about X" / "What's in my archive about [topic]?"

Workflow

1. User provides the trending topic or theme
2. Grep references/transcripts/ for the topic and related keywords (brainstorm 3-5 keyword variations)
3. Read the matching transcript sections — look for segments that are:
   • Self-contained (make sense without the full episode context)
   • Quotable (strong opinions, specific numbers, memorable phrasing)
   • Timeless (the insight still applies, not dated by a specific event that's now irrelevant)
4. For each clip candidate, extract the key passage and note which speaker said it

Output: Clip Candidates

For each candidate (aim for 5-10):

Episode: [ID / title] Speaker: [Sam / Shaan / Guest name] Key Passage:

"[Direct quote from transcript — the clip-worthy section, 2-5 sentences]"

Suggested Clip Title: [Punchy title for social media / YouTube Shorts] Why It Works Now: [1 sentence connecting this archive content to the user's trending topic]

Final Synthesis: Top 5 Recommendations

After completing any mode (or multiple modes), end with a ranked list:

Top 5 Content Opportunities

Rank by two factors:

1. Content potential — how compelling would this be as an episode/clip/post?
2. Archive evidence strength — how much source material exists to support it?

For the #1 recommendation, include a one-paragraph pitch: what the episode/content would be, why it works, and which archive material to pull from.

Deep-Dive References

These framework articles are rich content sources for mining:

• Read references/frameworks/personal-branding.md for content angles on personal brand building
• Read references/frameworks/creator-economy.md for creator business model evolution
• Read references/frameworks/content-to-commerce.md for audience-to-product pipeline stories
• Read references/frameworks/newsletter-business.md for newsletter-specific content angles
• Read references/frameworks/community-building.md for community-driven content strategies
• Read references/frameworks/my-first-million.md for meta-content about the show itself

Company/person profiles that generate repeat content:

• Read references/frameworks/acquisition-com.md for Hormozi empire updates and angles
• Read references/frameworks/morning-brew.md for [[frameworks/newsletter-business|newsletter business]] evolution
• Read references/frameworks/feastables.md for MrBeast commerce case study
• Read references/frameworks/hampton.md for peer community model updates
• Read references/frameworks/late-checkout.md for [[frameworks/greg-isenberg|Greg Isenberg]]'s community plays
• Read references/frameworks/skool.md for Hormozi's community platform

File Paths

All paths are relative to the plugin root (plugin/):