Skip to content

Instantly share code, notes, and snippets.

View jmanhype's full-sized avatar
🚒
Shipping

Straughter "BatmanOsama" Guthrie jmanhype

🚒
Shipping
View GitHub Profile
@jmanhype
jmanhype / generate_riley_assets.sh
Last active February 4, 2026 17:43
NANO BANANA PROTOCOL: Raw Bash Script for Consistent AI Character Generation
#!/bin/bash
# 🍌 NANO BANANA PRO: RAW ASSET GENERATION PROTOCOL
# Session: Feb 4, 2026
# Model: gemini-3-pro-image-preview
# Logic: Base64 Reference Injection + Prompt Engineering for Consistency
# 1. PRE-FLIGHT: ENCODE DNA
# Ensure 'riley.jpg' exists in the current directory.
if [ ! -f "riley.jpg" ]; then
@jmanhype
jmanhype / TOTAL_RECALL.md
Created February 4, 2026 17:42
TOTAL RECALL: Full Technical Dump of Riley AI Factory Session

🧠 TOTAL RECALL: The "Riley" AI Factory Full Session Dump

Session ID: Feb 4, 2026 Operator: Gemini CLI Objective: End-to-End Character Video Production & Distribution


πŸ—οΈ 1. INFRASTRUCTURE & REPAIR

The exact commands used to fix the 3090 environment.

@jmanhype
jmanhype / RILEY_FACTORY_GOD_MODE.md
Created February 4, 2026 17:40
THE SHIT: Full Data Manifest for Riley AI Video Factory

🧬 AI Video Factory: THE GOD MODE MANIFEST

Date: Feb 4, 2026 Target: Riley Guthrie Character Project


🧬 1. CHARACTER DNA (Reference Image Base64)

Use this in every Nano Pro prompt to maintain identity.

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
@jmanhype
jmanhype / PROJECT_ARTIFACTS.md
Created February 4, 2026 17:39
AI Video Factory: PROJECT ARTIFACTS (Riley Project)

πŸ“¦ Project Artifacts: Riley Character Story (Session Feb 4, 2026)

This manifest lists every specific asset generated, processed, and stored during this session.

πŸ–ΌοΈ 1. Source Assets (Nano Banana Pro)

Generated locally and uploaded to 3090 (~/ComfyUI/input/).

πŸ™οΈ NYC Date Night (V1)

  • riley_escalade.png - Luxury pickup shot
  • knicks_cheers.png - POV Toast shot
@jmanhype
jmanhype / factory_manifest_detailed.md
Last active February 4, 2026 18:01
AI Video Factory: The 'Riley' Protocol (Deep Dive SOP)

🏭 AI Video Factory: THE DEFINITIVE "RILEY" PROTOCOL (Technical Deep Dive)

Objective: End-to-end character consistency, video orchestration, and automated distribution. Hardware: Remote RTX 3090 (192.168.1.143) + Local Orchestrator. Session Date: Feb 4, 2026


🧬 1. CHARACTER DNA (The Identity Key)

This Base64 string must be injected into the inlineData of every Nano Pro call to maintain facial structure.

@jmanhype
jmanhype / factory_manifest.md
Created February 4, 2026 17:37
AI Video Factory: Full Production & Distribution Blueprint (Riley Project)

πŸš€ AI Video Factory: End-to-End Production & Distribution Manifest

This manifest documents the workflows, infrastructure, and automation logic developed for the Riley Character Story project. It is designed to be executed by a sub-agent or used as a blueprint for system replication.

πŸ–₯️ 1. Infrastructure (The 3090 Backend)

  • Host: [email protected]
  • Service: ComfyUI (Port 8188) using LTX-Video Models (FP8).
  • Orchestration: Letta MAS (Port 8283) with Director and Writer agents.
  • Control Plane: Comfy Pilot MCP Server (installed at ).
@jmanhype
jmanhype / x_article_automation.md
Created February 4, 2026 17:28
Automating X (Twitter) Article Publishing with agent-browser and Python

X Article Publisher Automation with agent-browser

This guide demonstrates how to automate publishing rich-text articles to X (Twitter) using agent-browser, Python, and the MacOS clipboard. It bypasses X's lack of an API for long-form articles by automating the web editor.

The Strategy

  1. Parse Markdown: Extract title, cover image, and convert body to HTML.
  2. Browser Automation: Open X editor, upload cover, and fill title.
  3. Rich Text Paste: Use Python to put HTML on the system clipboard as rich text, then paste (Cmd+V) into the browser.
  4. Inline Images: Copy images to clipboard and paste them at specific insertion points.
@jmanhype
jmanhype / facebook_upload_guide.md
Created February 4, 2026 16:56
Guide for automating Facebook video uploads with agent-browser

Automating Facebook Video Uploads with agent-browser

Facebook uses hidden <input type="file"> elements that are triggered by stylized UI components. To automate video uploads using the agent-browser CLI, follow this workflow.

1. Identify the Hidden Input

Facebook often has multiple hidden file inputs for different media types. Use eval to find the one that accepts video formats (look for video/* or specific extensions like matroska/mp4).

agent-browser eval "Array.from(document.querySelectorAll('input[type=file]')).map(i => i.accept)"
@jmanhype
jmanhype / OPENCLAW_OVERVIEW.md
Last active February 3, 2026 17:28 — forked from mikehostetler/OPENCLAW_OVERVIEW.md
OpenClaw Architecture Overview

OpenClaw Architecture Overview

OpenClaw is a multi-channel message routing layer wrapped around an embedded "coding agent" runtime (pi-agent-core). It enables AI agents to operate across messaging platforms (Telegram, Discord, Slack, Signal, iMessage, WhatsApp, etc.) with unified tool access, session management, and extensibility.


High-Level Architecture

β”Œβ”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”€β”
@jmanhype
jmanhype / factory_gist.md
Created January 30, 2026 23:53
20 Industries in 20 Seconds: The Autonomic Factory

20 Industries in 20 Seconds: The Autonomic Factory

Date: January 30, 2026 Achievement: Mass-Production of Industry-Specific SaaS Templates

The Breakthrough

We have moved beyond "coding assistants" to an Autonomic Software Factory.

In a single autonomous run (Item 100), the Wreckit agent:

  1. Researched the top 20 "Long Tail" industry verticals for the GCP Marketplace.
  2. Designed a standardized "80/20" architecture (80% common Erlang/OTP core, 20% domain specific).