@opendirectory.dev/skills 0.1.65 → 0.1.66

package/skills/linkedin-post-generator/README.md

@@ -13,7 +13,7 @@ npx "@opendirectory.dev/skills" install linkedin-post-generator --target claude
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -27,3 +27,105 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+## Post Styles
+
+| Style | Use When | Example Input |
+|-------|----------|---------------|
+| Founder/Ship | Personal story of building or shipping | "We merged our streaming SDK after 3 weeks" |
+| Insight | Educational observation or lesson | Blog article, pattern you noticed, lesson learned |
+| Product Launch | Announcing a new tool or feature | PR description, launch brief, feature going GA |
+| Tutorial Summary | Distilling a long technical post | Tutorial URL, deep-dive article, step-by-step guide |
+
+The agent auto-detects the right style. You override it by asking for a specific one.
+
+## Requirements
+
+No LLM API key needed. The agent writes the post.
+
+Composio is optional. Add it to post directly to LinkedIn.
+
+## Setup
+
+### Composio (Optional)
+
+Without Composio, the agent outputs formatted text for copy-paste. No configuration needed.
+
+To enable direct posting:
+1. Get your API key at https://app.composio.dev/settings
+2. Connect your LinkedIn account at https://app.composio.dev/app/linkedin
+3. Complete the OAuth flow
+4. Add the key to your .env file:
+
+```bash
+cp .env.example .env
+# Edit .env and add your COMPOSIO_API_KEY
+```
+
+## How to Use
+
+From a URL:
+```
+"Turn this into a LinkedIn post: https://yourblog.com/my-post"
+"Write a LinkedIn post about this article: [URL]"
+```
+
+From pasted text:
+```
+"Here's a case study, turn it into a LinkedIn post: [paste text]"
+"Convert this to a LinkedIn post: [paste article]"
+```
+
+From a PR or shipped feature:
+```
+"Write a LinkedIn post about this PR we merged: [paste PR description]"
+"Announce our new feature on LinkedIn: [describe the feature]"
+```
+
+With a style override:
+```
+"Write a LinkedIn post in Tutorial Summary style about: [topic]"
+"Use the Product Launch style for this: [description]"
+```
+
+With direct posting:
+```
+"Post this to LinkedIn: [paste content]"
+"Generate and post a LinkedIn update about [topic]"
+```
+
+## Output
+
+| Output | Description |
+|--------|-------------|
+| LinkedIn post | Formatted text, 900-1,300 characters, ready to publish |
+| First comment | Text with source links. Post this immediately after publishing. |
+| Hook alternatives | 2 additional hook lines in different formats |
+| Posted confirmation | If Composio is configured and you confirm posting |
+
+## How Posts Are Formatted
+
+Four rules drive every post the agent writes:
+
+- Hook first. The first line is all most people see. It works standalone.
+- Short paragraphs. 1-3 lines, then a blank line. LinkedIn is mobile-first.
+- Links in the first comment. URLs in the post body reduce LinkedIn's distribution.
+- Question or CTA at the end. One or the other, not both.
+
+## Project Structure
+
+```
+linkedin-post-generator/
+├── SKILL.md
+├── README.md
+├── .env.example
+├── evals/
+│   └── evals.json
+└── references/
+    ├── linkedin-format.md
+    └── output-template.md
+```
+
+## License
+
+MIT

package/skills/llms-txt-generator/README.md

@@ -13,7 +13,7 @@ npx "@opendirectory.dev/skills" install llms-txt-generator --target claude
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -27,3 +27,140 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+## What It Does
+
+The skill crawls your website using Chrome DevTools, reads your actual pages, and produces a clean `llms.txt` file in the format specified by [Jeremy Howard's llms.txt standard](https://llmstxt.org). When AI agents (Claude, ChatGPT, Gemini) visit your site, they read `llms.txt` first to understand what you are and where to find authoritative content.
+
+**Without llms.txt:** AI agents guess, hallucinate, or cite competitors instead.  
+**With llms.txt:** AI agents cite your product correctly and know exactly where your docs, blog, and key pages live.
+
+## Two Modes
+
+### Codebase Mode (no Chrome needed)
+If you're inside a website's repo, the skill reads your source files directly pages, routes, blog posts, frontmatter, site config. It writes `llms.txt` straight to `public/` when you approve. No browser required.
+
+Supported frameworks: **Next.js** (pages + app router), **Astro**, **Nuxt**, **Gatsby**, **SvelteKit**, **Hugo**, **Jekyll**
+
+### Live Site Mode (Chrome or fetch fallback)
+If you only have the URL, the skill crawls the live site using Chrome DevTools MCP. Falls back to standard web fetch if Chrome isn't available.
+
+---
+
+## Requirements
+
+**Codebase Mode:** No extra setup. Just be inside the repo directory.
+
+**Live Site Mode:**
+- Chrome with remote debugging enabled (or any live URL skill will fall back to web fetch)
+- Chrome DevTools MCP server configured in your agent (optional, improves JS-rendered sites)
+
+## Setup
+
+### For Live Site Mode: Start Chrome with Remote Debugging
+
+**Mac:**
+```bash
+open -a "Google Chrome" --args --remote-debugging-port=9222
+```
+
+**Linux:**
+```bash
+google-chrome --remote-debugging-port=9222
+```
+
+**Windows:**
+```bash
+chrome.exe --remote-debugging-port=9222
+```
+
+### 2. Install Chrome DevTools MCP Server
+
+Follow the setup at: https://github.com/ChromeDevTools/chrome-devtools-mcp
+
+Add to your agent's MCP configuration:
+```json
+{
+  "mcpServers": {
+    "chrome-devtools": {
+      "command": "npx",
+      "args": ["-y", "@chrome-devtools/mcp-server"],
+      "env": {
+        "CHROME_DEBUGGING_PORT": "9222"
+      }
+    }
+  }
+}
+```
+
+### 3. Configure Environment (Optional)
+
+Copy `.env.example` to `.env` and fill in:
+```bash
+cp .env.example .env
+```
+
+`GITHUB_TOKEN` and `GITHUB_REPO` are only needed if you want the agent to automatically open a GitHub PR with the generated file.
+
+## How to Use
+
+**Codebase Mode** just be inside your project and ask:
+```
+"Generate an llms.txt for this site"
+"Add llms.txt to this project"
+"Make this site readable by AI agents"
+```
+
+The agent will detect your framework, read your pages and blog posts from source, generate `llms.txt`, and write it to the right directory (e.g. `public/llms.txt`) after you confirm.
+
+**Live Site Mode** provide a URL:
+```
+"Generate an llms.txt for https://yoursite.com"
+"Does https://yoursite.com have an llms.txt? If not, create one."
+```
+
+The agent will:
+1. Check if `llms.txt` already exists at the domain
+2. Crawl homepage, docs, blog, about, pricing, and API pages
+3. Generate `llms.txt` following the official spec
+4. Optionally generate `llms-full.txt` with full page content
+5. Save the file locally and give you deployment instructions
+6. Optionally open a GitHub PR if configured
+
+## Where to Deploy the File
+
+Place `llms.txt` at your web root so it's accessible at `https://yourdomain.com/llms.txt`:
+
+| Platform | File Location |
+|----------|--------------|
+| Next.js / Vercel | `/public/llms.txt` |
+| Astro | `/public/llms.txt` |
+| Nuxt | `/public/llms.txt` |
+| GitHub Pages | Repository root |
+| Hugo | `/static/llms.txt` |
+| WordPress | Upload via FTP to web root |
+
+## Output Files
+
+| File | Description |
+|------|-------------|
+| `llms.txt` | Structured link map. LLMs follow links to find content |
+| `llms-full.txt` | Full prose content of key pages. LLMs ingest everything at once |
+
+## Project Structure
+
+```
+llms-txt-generator/
+├── SKILL.md                          # Agent instructions
+├── README.md                         # This file
+├── .env.example                      # Environment variables template
+├── evals/
+│   └── evals.json                    # Test prompts for skill evaluation
+└── references/
+    ├── llms-txt-spec.md              # The llms.txt format specification
+    └── output-template.md            # Exact output template with example
+```
+
+## License
+
+MIT

package/skills/luma-attendees-scraper/README.md

@@ -168,24 +168,3 @@ Some browsers block downloads from DevTools-triggered scripts until user interac
 ## Disclaimer
 
 Use this responsibly and only for events and attendee lists your account is authorized to access.
-
-
-## Install
-
-### Video Tutorial
-Watch this quick video to see how it's done:
-
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
-
-### Step 1: Download the skill from GitHub
-1. Copy the URL of this specific skill folder from your browser's address bar.
-2. Go to [download-directory.github.io](https://download-directory.github.io/).
-3. Paste the URL and click **Enter** to download.
-
-### Step 2: Install the Skill in Claude
-1. Open your **Claude desktop app**.
-2. Go to the sidebar on the left side and click on the **Customize** section.
-3. Click on the **Skills** tab, then click on the **+** (plus) icon button to create a new skill.
-4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
-
-> **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!

package/skills/map-your-market/README.md

@@ -11,7 +11,7 @@ npx "@opendirectory.dev/skills" install map-your-market --target claude
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -25,3 +25,123 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+## What It Does
+
+- Accepts any combination of: product description, category keywords, competitor names
+- Auto-detects relevant subreddits from the category
+- Searches Reddit public JSON API for pain posts (top posts, last 12 months)
+- Searches Hacker News Algolia API for stories and Ask HN threads
+- Searches GitHub Issues on competitor repos for high-reaction complaints
+- Scrapes G2 category pages for vendor count and top products
+- Fetches Google Trends direction (up / flat / down) for the category
+- Scores every signal by source weight and recency (GitHub issues score 3x Reddit -- more deliberate signal)
+- Clusters top 60 signals into 5-7 named pain themes
+- Extracts ICP from subreddit and post metadata (not just content)
+- Generates a positioning framework with messaging angles using verbatim market language
+- Saves output to `docs/market-maps/[category]-[date].md` + JSON snapshot
+
+## Requirements
+
+| Requirement | Purpose | How to Set Up |
+|---|---|---|
+| GITHUB_TOKEN | Optional -- improves GitHub Issues rate limit from 60/hr to 5000/hr | github.com/settings/tokens (no scopes needed for public repos) |
+
+No other API keys required.
+
+## Setup
+
+```bash
+cp .env.example .env
+# Add GITHUB_TOKEN if you want higher GitHub rate limits
+```
+
+## How to Use
+
+```
+"Map my market: I build developer observability tools"
+"Who is my ICP? Competitors: Datadog, Grafana, New Relic"
+"What are the top pains in the HR software market?"
+"Find messaging angles for my B2B analytics tool"
+"Map the CRM market. What are people complaining about?"
+"I build payment APIs. Who should I be selling to?"
+```
+
+Include competitor names for richer GitHub Issues data. Include a product description for tailored messaging angles.
+
+## Why This Instead of Manual Research
+
+A founder doing this manually would spend 2-3 days:
+- Reading Reddit threads, taking notes
+- Scrolling HN "Ask HN" posts
+- Checking G2 review counts per vendor
+- Looking up Google Trends
+- Synthesizing into a document
+
+This skill does the same sweep in 3 minutes and returns verbatim quotes, not paraphrased summaries. The messaging framework uses the exact language your market uses -- not marketing copy you invented.
+
+## The Pain Score
+
+`pain_score = base * recency_factor`
+
+- GitHub issue reactions: `reactions * 3` -- a developer deliberately clicking +1 is the strongest signal
+- Reddit: `upvotes + (comments * 0.3)` -- upvotes count more than comments (comments include noise)
+- HN: `points + (comments * 0.3)` -- same structure
+
+Score tiers: critical (200+), high (50-199), medium (10-49), noise (<10, filtered out).
+
+## Velocity Tracking
+
+Run the skill every quarter. JSON snapshots in `docs/market-maps/` let you compare pain cluster rankings over time. A pain that was #3 last quarter and is #1 this quarter is accelerating -- a stronger positioning bet.
+
+## Cost Per Run
+
+- Reddit, HN, Google Trends: free, no auth
+- GitHub Issues: free with optional token
+- G2 scrape: free HTML fetch
+- AI analysis: uses the model already running the skill
+- Total: free
+
+## Standalone Script
+
+Run data collection without Claude. Useful when you want the raw signals first, then bring them to any AI for analysis.
+
+```bash
+# Basic usage
+python3 scripts/fetch.py "developer observability"
+
+# With competitors
+python3 scripts/fetch.py "developer observability" --competitors "Datadog,Grafana,New Relic"
+
+# With product context
+python3 scripts/fetch.py "B2B analytics" --context "We help ops teams track spend"
+
+# Print to stdout
+python3 scripts/fetch.py "devops tooling" --stdout | jq '.summary'
+
+# With GitHub token for higher rate limits
+GITHUB_TOKEN=your_token python3 scripts/fetch.py "CRM software" --competitors "Salesforce,HubSpot" --output results.json
+```
+
+The script writes a JSON file with all raw signals. Open that file with Claude and ask: "Generate a market map and positioning framework from this data."
+
+## Project Structure
+
+```
+map-your-market/
+├── SKILL.md
+├── README.md
+├── .env.example
+├── scripts/
+│   └── fetch.py           standalone data collector
+├── evals/
+│   └── evals.json
+└── references/
+    ├── subreddit-map.md   category to subreddit mapping
+    ├── pain-scoring.md    scoring formula and tier thresholds
+    └── icp-signals.md     how to extract ICP from post metadata
+```
+
+## License
+
+MIT

package/skills/meeting-brief-generator/README.md

@@ -14,7 +14,7 @@ npx "@opendirectory.dev/skills" install meeting-brief-generator --target claude
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -28,3 +28,87 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+## What It Does
+
+- Runs 6-8 targeted Tavily searches covering company overview, recent news, tech stack, product, competitors, funding, and contact background
+- Synthesizes results into a structured 1-page brief using Gemini
+- Every claim cites a source URL from the research
+- Optionally saves the brief to a Notion database
+- Handles low-data companies by marking gaps instead of inventing content
+
+## Requirements
+
+| Requirement | Purpose | How to Set Up |
+|------------|---------|--------------|
+| Tavily API key | Company research | app.tavily.com, API Keys |
+| Gemini API key | Brief synthesis | aistudio.google.com, Get API key |
+| Notion token (optional) | Saving briefs | notion.so/my-integrations |
+
+## Setup
+
+```bash
+cp .env.example .env
+```
+
+Fill in:
+- `TAVILY_API_KEY` (required)
+- `GEMINI_API_KEY` (required)
+- `NOTION_TOKEN` and `NOTION_DATABASE_ID` (optional, for saving briefs)
+
+## How to Use
+
+Basic brief with company only:
+
+```
+"Prepare me for a meeting with Stripe next Tuesday"
+"Generate a meeting brief for Vercel"
+"Research Acme Corp before my call tomorrow"
+```
+
+With contact and meeting type:
+
+```
+"Prepare a discovery call brief for Linear. I'm meeting with the VP Engineering, Jordan Lee."
+"Create a pre-call brief for Notion. Demo call on April 20."
+```
+
+Save to Notion:
+
+```
+"Generate a meeting brief for Figma and save it to Notion"
+```
+
+## Brief Sections
+
+| Section | Content |
+|---------|---------|
+| Company Snapshot | What they do, size, funding stage, HQ |
+| Recent News | Last 30 days, source URLs |
+| Decision Maker | Name, title, background (if contact provided) |
+| Tech Stack Signals | Tools spotted in job posts, blog, GitHub |
+| Competitive Context | Who they compete with and how |
+| Talking Points | Because/mention/to formula, 3-5 bullets |
+| Open Questions | Company-specific discovery questions |
+
+## Output Format
+
+One page, under 400 words. Every claim has a source URL. Talking points follow the format: "Because [finding from research], mention [point] to [goal]."
+
+## Project Structure
+
+```
+meeting-brief-generator/
+├── SKILL.md
+├── README.md
+├── .env.example
+├── evals/
+│   └── evals.json
+└── references/
+    ├── brief-format.md
+    └── output-template.md
+```
+
+## License
+
+MIT

package/skills/meta-ads-skill/README.md

@@ -11,7 +11,7 @@ npx "@opendirectory.dev/skills" install meta-ads-skill --target claude
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -25,3 +25,69 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+## Overview
+
+The **Meta Ads Skill** is a comprehensive, production-ready  skill designed to give LLMs and AI agents expert-level capabilities to orchestrate the official **Meta Ads Python CLI**.
+
+By using this skill, an agent transforms into an **Expert Media Buyer**. It will know exactly how to explore ad structures, troubleshoot campaign performance (like CPA spikes), discover new audiences, and format massive Meta APIs JSON payloads into beautiful, readable markdown reports.
+
+---
+
+## For Agents: How to Use This Skill Efficiently
+
+This skill is designed using a **Progressive Disclosure (Hub-and-Spoke)** architecture to maximize context window efficiency:
+
+1. **The Hub (`SKILL.md`)**: The primary entry point. It provides strict guardrails, safety protocols, and the authentication troubleshooting workflow.
+2. **The Spokes (`references/`)**: 
+   - When you need to perform a specific task (e.g., investigating a CPA spike), read `references/workflows.md` for the exact step-by-step orchestration strategy.
+   - When presenting data to the user, read `references/report_templates.md` to strictly follow the required Markdown layout.
+
+###  Strict Agent Guardrails
+* **Context Protection**: ALWAYS default to `time_range="last_7d"` for insights. ALWAYS use `limit=10` for listing campaigns/adsets initially.
+* **Safety First**: NEVER execute state-changing tools (`create_campaign`, `update_campaign`) without explicitly showing the parameters to the user and waiting for their affirmative confirmation.
+
+---
+
+## Installation & Setup
+
+To use this skill, you must install the official Meta Ads CLI and configure your credentials.
+
+### 1. Install the CLI
+The skill relies on the `meta-ads` Python package:
+```bash
+pip install meta-ads
+```
+
+### 2. Authentication (System User Token)
+The CLI uses a **System User Access Token** for authentication. 
+1. Generate a System User Token in your [Meta Business Suite](https://business.facebook.com/settings/system-users).
+2. Ensure the token has `ads_management`, `ads_read`, and `read_insights` permissions.
+3. Set the following environment variables on your machine:
+
+```bash
+export ACCESS_TOKEN="your_system_user_access_token"
+export AD_ACCOUNT_ID="act_your_ad_account_id"
+```
+
+---
+
+## Skill Repository Structure
+
+When you deploy this skill, the structure will look like this:
+
+```text
+meta-ads-skill/
+ SKILL.md                          # The core router & guardrails
+ references/
+    report_templates.md           # Standardized markdown report structures
+    workflows.md                  # Orchestration strategies (e.g., CPA troubleshooting)
+```
+
+## Supported Commands
+
+This skill orchestrates the `meta-ads` CLI using a noun-verb structure:
+* **Campaigns**: `meta ads campaign list`, `meta ads campaign create`
+* **Ad Sets**: `meta ads adset list`
+* **Ads**: `meta ads ad list`
+* **Insights**: `meta ads insights get`

package/skills/meta-tribeV2-skill/README.md

@@ -21,12 +21,10 @@ This skill provides the infrastructure to host the massive 80GB TRIBE v2 model p
 
 ## Install
 
-## Install
-
 ### Video Tutorial
 Watch this quick video to see how it's done:
 
-https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
+https://github.com/user-attachments/assets/ee98a1b5-ebc4-452f-bbfb-c434f2935067
 
 ### Step 1: Download the skill from GitHub
 1. Copy the URL of this specific skill folder from your browser's address bar.
@@ -40,3 +38,66 @@ https://github.com/user-attachments/assets/cea8b565-2002-4a87-8857-d902bfcfdc1c
 4. Choose the option to **Upload a skill**, and drag and drop the `.zip` file (or you can extract it and drop the folder, both work).
 
 > **Note:** For some skills (like `position-me`), the `SKILL.md` file might be located inside a subfolder. Always make sure you are uploading the specific folder that contains the `SKILL.md` file!
+
+---
+
+## Deployment Options
+
+Because TRIBE v2 requires a massive amount of VRAM (24GB for text, up to 80GB for video), we offer 3 different deployment options so anyone can use it, regardless of budget or technical expertise.
+
+### 1. Google Colab (Zero Cost, Decoupled)
+Best for users without a cloud budget. Colab provides free T4 GPUs.
+* How it works: We use a decoupled architecture. You run the heavy AI inference in a Colab Notebook, which outputs a preds.npy prediction file. You then run a local script on your laptop to generate the report.
+* Setup:
+  1. Open Google Colab and upload the script from scripts/colab_inference.py into a new Notebook.
+  2. Run the notebook. It will output preds.npy and segments.json.
+  3. Download those files to your machine and run: `python scripts/local_analyze.py --preds preds.npy`. This will output a text report and an ASCII terminal graph showing the engagement peaks and valleys.
+
+### 2. RunPod (Serverless, Pay-per-second)
+Best for production agents and developers. You only pay for the seconds the model is running.
+* How it works: We provide a RunPod Handler and a custom Dockerfile that caches the 80GB model inside the image.
+* Setup:
+  1. Build the Docker image using server/Dockerfile.runpod: docker build -f Dockerfile.runpod -t tribe-runpod .
+  2. Push the image to Docker Hub or GHCR.
+  3. Create a new RunPod Serverless Endpoint using your image URL.
+  4. Point your AI Agent to your RunPod Endpoint URL.
+
+### 3. AWS EC2 Persistent (Enterprise, BYO-Compute)
+Best for heavy, continuous usage. 
+* How it works: Automatically provisions an AWS g5.12xlarge instance (4x A10G GPUs) and runs a FastAPI server.
+* Setup:
+  1. Ensure your AWS account has a vCPU quota limit of at least 48 for "Running On-Demand G and VT instances".
+  2. Run bash scripts/launch_persistent.sh to provision the instance.
+  3. Run export HF_TOKEN="your_token" followed by bash scripts/deploy_to_persistent.sh to build and launch the Docker API.
+
+#### AWS GPU Lifecycle & Estimated Costs
+Running the `g5.12xlarge` instance (4x A10G GPUs) provides incredible speed but costs **$7.09 per hour** on On-Demand pricing. It is crucial to manage this lifecycle.
+1. **Launch:** Run `bash scripts/launch_persistent.sh` (Takes ~3 minutes).
+2. **Analyze:** Run your videos through the API.
+3. **Terminate:** When you are completely finished for the day, you MUST terminate the instance to stop billing.
+   - Run `aws ec2 describe-instances --filters "Name=instance-state-name,Values=running"` to find your Instance ID.
+   - Run `aws ec2 terminate-instances --instance-ids <YOUR_INSTANCE_ID>`.
+   - *Do not just "stop" the instance if you don't want to pay for EBS Volume storage costs overnight. Terminate it.*
+
+---
+
+## HuggingFace Authentication (Required for all methods)
+
+TRIBE v2 relies on meta-llama/Llama-3.2-3B, which is a Gated Model.
+1. Create a HuggingFace account.
+2. Go to the Llama 3.2 3B page and TRIBE v2 page and agree to Meta's license terms.
+3. Generate a HuggingFace Access Token (Read permissions) at huggingface.co/settings/tokens.
+4. Supply this token via the HF_TOKEN environment variable.
+
+---
+
+## The Neuroscience of the Engagement Report
+
+The AI agent will read the raw API output and translate the neuroscience into plain English for you:
+
+* VAN (Ventral Attention Network): Translated to "Is this surprising enough to stop a scroll?". High VAN means the content is novel and creates a pattern interrupt.
+* DMN (Default Mode Network): Translated to "Will people get bored and tune out?". High DMN is bad. It means the brain is wandering. The AI uses this to identify "Cut Candidates" in your video.
+* DAN (Dorsal Attention Network): Translated to "Are people actively following along?". High DAN means strong logical focus.
+* Limbic Network: Translated to "Does this make people feel something?". High Limbic means strong emotional response.
+
+Check out the [Results Showcase](results_showcase.md) for actual examples of Neuro-Marketing reports generated by this skill.