# timsonner/agent-skills architecture

> Architecture diagram of the GitHub repository timsonner/agent-skills, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-06). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/timsonner/agent-skills
- Repository: https://github.com/timsonner/agent-skills

## Overview

This repository is a catalog of reusable agent skills, operating guides, and a few task-specific scripts—not one integrated runtime. Its clearest end-to-end workflow is the workstation-drive loop: an agent loads the skill, configures a harness if needed, connects to a supplied remote-desktop environment, then interacts through screenshots and UI injection. Other branches cover coding-agent harnesses, platform guidance, administration, security, and PDF-to-Tinker training. The README and sampled files support the relationships shown; many unsampled skills are represented by their documented purpose, without assuming internal wiring.

Main entry points: [`agent-workstation-drive/SKILL.md`](https://github.com/timsonner/agent-skills/blob/main/agent-workstation-drive/SKILL.md) (Workstation Skill), [`windows-ui-inject/scripts/Invoke-MstscConnect.ps1`](https://github.com/timsonner/agent-skills/blob/main/windows-ui-inject/scripts/Invoke-MstscConnect.ps1) (MSTSC Connector), [`pdf-to-tinker-training/scripts/to_datum.py`](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/to_datum.py) (Datum Builder), [`pdf-to-tinker-training/scripts/mock_tokenizer.py`](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/mock_tokenizer.py) (Tokenizer Workaround)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workstation["Workstation Drive"]
  node_drive_skill["Workstation Skill<br/>[SKILL.md]"]
  node_harness_install["Harness Setup"]
  node_rdp_server["Linux RDP Skills"]
  node_rdp_client["Windows UI Inject<br/>[SKILL.md]"]
  node_mstsc["MSTSC Connector"]
end

subgraph group_harnesses["Agent Harnesses"]
  node_harness_guides["CLI Harness Guides<br/>[SKILL.md]"]
  node_offload["Harness Offload"]
  node_copilot_sdk["Copilot SDK<br/>[SKILL.md]"]
  node_mcp["MCP Server Guide<br/>[SKILL.md]"]
end

subgraph group_platform["Platform Skills"]
  node_opencode_guides["OpenCode Guides<br/>[SKILL.md]"]
  node_admin_guides["Admin Guides<br/>[SKILL.md]"]
  node_container_guides["Container Guides<br/>[SKILL.md]"]
  node_desktop_guides["Desktop Driving<br/>[SKILL.md]"]
  node_browser_guides["Browser Helpers<br/>[SKILL.md]"]
end

subgraph group_security["Security Skills"]
  node_pentest_guides["Pentest Skills<br/>[SKILL.md]"]
end

subgraph group_training["Training Data"]
  node_pdf_chunks["PDF Chunking<br/>[chunk_pdf.py]"]
  node_mock_tokenizer["Tokenizer Workaround<br/>[mock_tokenizer.py]"]
  node_datum_builder["Datum Builder<br/>[to_datum.py]"]
  node_lora_trainer["LoRA Trainer<br/>[train_lora.py]"]
end

node_agent(("AI Agent"))
node_remote_desktop(("Remote Desktop"))

node_agent -->|"loads"| node_drive_skill
node_drive_skill -->|"sets up"| node_harness_install
node_drive_skill -->|"uses server skill"| node_rdp_server
node_drive_skill -->|"connects with"| node_mstsc
node_mstsc -->|"opens session"| node_remote_desktop
node_agent -->|"drives session"| node_rdp_client
node_agent -->|"loads"| node_harness_guides
node_agent -->|"delegates via"| node_offload
node_agent -->|"integrates with"| node_copilot_sdk
node_agent -->|"builds with"| node_mcp
node_agent -->|"uses"| node_opencode_guides
node_agent -->|"uses"| node_admin_guides
node_agent -->|"uses"| node_container_guides
node_agent -->|"uses"| node_desktop_guides
node_agent -->|"uses"| node_browser_guides
node_agent -->|"uses"| node_pentest_guides
node_pdf_chunks -->|"chunks feed"| node_datum_builder
node_datum_builder -->|"imports workaround"| node_mock_tokenizer
node_lora_trainer -->|"imports workaround"| node_mock_tokenizer

click node_drive_skill "https://github.com/timsonner/agent-skills/blob/main/agent-workstation-drive/SKILL.md"
click node_harness_install "https://github.com/timsonner/agent-skills/blob/main/agent-workstation-drive/scripts/Install-ToHarness.ps1"
click node_rdp_server "https://github.com/timsonner/agent-skills/tree/main/gnome-wayland-rdp"
click node_rdp_client "https://github.com/timsonner/agent-skills/blob/main/windows-ui-inject/SKILL.md"
click node_mstsc "https://github.com/timsonner/agent-skills/blob/main/windows-ui-inject/scripts/Invoke-MstscConnect.ps1"
click node_harness_guides "https://github.com/timsonner/agent-skills/blob/main/agy-harness/SKILL.md"
click node_offload "https://github.com/timsonner/agent-skills/blob/main/harness-offload/scripts/Invoke-HarnessOffload.ps1"
click node_copilot_sdk "https://github.com/timsonner/agent-skills/blob/main/copilot-sdk/SKILL.md"
click node_mcp "https://github.com/timsonner/agent-skills/blob/main/mcp-server/SKILL.md"
click node_opencode_guides "https://github.com/timsonner/agent-skills/blob/main/opencode/intro/SKILL.md"
click node_admin_guides "https://github.com/timsonner/agent-skills/blob/main/proxmox-system-administration/SKILL.md"
click node_container_guides "https://github.com/timsonner/agent-skills/blob/main/wsl-containers/SKILL.md"
click node_desktop_guides "https://github.com/timsonner/agent-skills/blob/main/drive-hyprland/SKILL.md"
click node_browser_guides "https://github.com/timsonner/agent-skills/blob/main/read-canvas-browser/SKILL.md"
click node_pentest_guides "https://github.com/timsonner/agent-skills/blob/main/pentest/methodology/SKILL.md"
click node_pdf_chunks "https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/chunk_pdf.py"
click node_mock_tokenizer "https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/mock_tokenizer.py"
click node_datum_builder "https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/to_datum.py"
click node_lora_trainer "https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/train_lora.py"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_drive_skill,node_harness_install,node_rdp_server,node_rdp_client,node_mstsc toneBlue
class node_harness_guides,node_offload,node_copilot_sdk,node_mcp toneAmber
class node_opencode_guides,node_admin_guides,node_container_guides,node_desktop_guides,node_browser_guides toneMint
class node_pentest_guides toneRose
class node_pdf_chunks,node_mock_tokenizer,node_datum_builder,node_lora_trainer,node_agent,node_remote_desktop toneIndigo
```

## Components

### Workstation Drive

- [Workstation Skill](https://github.com/timsonner/agent-skills/blob/main/agent-workstation-drive/SKILL.md): `agent-workstation-drive/SKILL.md`
- [Harness Setup](https://github.com/timsonner/agent-skills/blob/main/agent-workstation-drive/scripts/Install-ToHarness.ps1): `agent-workstation-drive/scripts/Install-ToHarness.ps1`
- [Linux RDP Skills](https://github.com/timsonner/agent-skills/tree/main/gnome-wayland-rdp): `gnome-wayland-rdp`
- [Windows UI Inject](https://github.com/timsonner/agent-skills/blob/main/windows-ui-inject/SKILL.md): `windows-ui-inject/SKILL.md`
- [MSTSC Connector](https://github.com/timsonner/agent-skills/blob/main/windows-ui-inject/scripts/Invoke-MstscConnect.ps1): `windows-ui-inject/scripts/Invoke-MstscConnect.ps1`

### Agent Harnesses

- [CLI Harness Guides](https://github.com/timsonner/agent-skills/blob/main/agy-harness/SKILL.md): `agy-harness/SKILL.md`
- [Harness Offload](https://github.com/timsonner/agent-skills/blob/main/harness-offload/scripts/Invoke-HarnessOffload.ps1): `harness-offload/scripts/Invoke-HarnessOffload.ps1`
- [Copilot SDK](https://github.com/timsonner/agent-skills/blob/main/copilot-sdk/SKILL.md): `copilot-sdk/SKILL.md`
- [MCP Server Guide](https://github.com/timsonner/agent-skills/blob/main/mcp-server/SKILL.md): `mcp-server/SKILL.md`

### Platform Skills

- [OpenCode Guides](https://github.com/timsonner/agent-skills/blob/main/opencode/intro/SKILL.md): `opencode/intro/SKILL.md`
- [Admin Guides](https://github.com/timsonner/agent-skills/blob/main/proxmox-system-administration/SKILL.md): `proxmox-system-administration/SKILL.md`
- [Container Guides](https://github.com/timsonner/agent-skills/blob/main/wsl-containers/SKILL.md): `wsl-containers/SKILL.md`
- [Desktop Driving](https://github.com/timsonner/agent-skills/blob/main/drive-hyprland/SKILL.md): `drive-hyprland/SKILL.md`
- [Browser Helpers](https://github.com/timsonner/agent-skills/blob/main/read-canvas-browser/SKILL.md): `read-canvas-browser/SKILL.md`

### Security Skills

- [Pentest Skills](https://github.com/timsonner/agent-skills/blob/main/pentest/methodology/SKILL.md): `pentest/methodology/SKILL.md`

### Training Data

- [PDF Chunking](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/chunk_pdf.py): `pdf-to-tinker-training/scripts/chunk_pdf.py`
- [Tokenizer Workaround](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/mock_tokenizer.py): `pdf-to-tinker-training/scripts/mock_tokenizer.py`
- [Datum Builder](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/to_datum.py): `pdf-to-tinker-training/scripts/to_datum.py`
- [LoRA Trainer](https://github.com/timsonner/agent-skills/blob/main/pdf-to-tinker-training/scripts/train_lora.py): `pdf-to-tinker-training/scripts/train_lora.py`

### Other components

- **AI Agent**
- **Remote Desktop**

## Connections

- AI Agent → Workstation Skill: loads (evidence: `README.md`)
- Workstation Skill → Harness Setup: sets up (evidence: `README.md`)
- Workstation Skill → Linux RDP Skills: uses server skill (evidence: `README.md`)
- Workstation Skill → MSTSC Connector: connects with (evidence: `README.md`)
- MSTSC Connector → Remote Desktop: opens session (evidence: `README.md`)
- AI Agent → Windows UI Inject: drives session (evidence: `README.md`)
- AI Agent → CLI Harness Guides: loads (evidence: `README.md`)
- AI Agent → Harness Offload: delegates via (evidence: `README.md`)
- AI Agent → Copilot SDK: integrates with (evidence: `README.md`)
- AI Agent → MCP Server Guide: builds with (evidence: `README.md`)
- AI Agent → OpenCode Guides: uses (evidence: `README.md`)
- AI Agent → Admin Guides: uses (evidence: `README.md`)
- AI Agent → Container Guides: uses (evidence: `README.md`)
- AI Agent → Desktop Driving: uses (evidence: `README.md`)
- AI Agent → Browser Helpers: uses (evidence: `README.md`)
- AI Agent → Pentest Skills: uses (evidence: `README.md`)
- PDF Chunking → Datum Builder: chunks feed (evidence: `pdf-to-tinker-training/scripts/to_datum.py`)
- Datum Builder → Tokenizer Workaround: imports workaround (evidence: `pdf-to-tinker-training/scripts/to_datum.py`)
- LoRA Trainer → Tokenizer Workaround: imports workaround (evidence: `pdf-to-tinker-training/scripts/train_lora.py`)
