A command-line workflow for controlling a local ComfyUI image-rendering server and keeping a lasting record of its runs. ComfyUI is a visual system for creating images from connected processing steps, and normally loses its run history when restarted.

Use it to submit ComfyUI graphs, check which models or loader inputs are available, find out how long a render took, and identify what produced an output file.

It provides a durable record when the rendering server’s own history disappears after a restart. It also helps inspect render details and troubleshoot models, loaders, and output files.

pp-comfyui — Skill by dmmdea

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add skills/dmmdea/offload-harness/comfyui

Any agent

npx skills add dmmdea/offload-harness --skill comfyui

Clone the repo

git clone --depth 1 https://github.com/dmmdea/offload-harness

Made for: Claude Code, Codex.

Per session 81 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 7,236 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00081	$0.07236
Opus 5	$0.00041	$0.03618
Sonnet 5	$0.00016	$0.01447
Haiku 4.5	$0.00008	$0.00724

Measured 2d ago against content hash ccf81510bb59, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

pp-comfyui scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

tools/comfyui/SKILL.md · 569 lines

How it starts

The opening of the file, as written. The whole thing — 569 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ComfyUI — Printing Press CLI

Prerequisites: Install the CLI

This skill drives the comfyui-pp-cli binary. You must verify the CLI is installed before invoking any command from this skill. If it is missing, install it first:

Install via the Printing Press installer. It defaults binaries to $HOME/.local/bin on macOS/Linux and %LOCALAPPDATA%\Programs\PrintingPress\bin on Windows:
```
npx -y @mvanhorn/printing-press-library install comfyui --cli-only
```
Verify: comfyui-pp-cli --version
Ensure the reported install directory is on $PATH for the agent/runtime that will invoke this skill.

If the npx install fails (no Node, offline, etc.), fall back to a direct Go install (requires Go 1.26.6 or newer). This installs into $GOPATH/bin (default $HOME/go/bin), so add that directory to $PATH instead:

go install github.com/mvanhorn/printing-press-library/library/ai/comfyui/cmd/comfyui-pp-cli@latest

If --version reports "command not found" after install, the runtime cannot see the binary directory on $PATH. Do not proceed with skill commands until verification succeeds.

ComfyUI keeps its render history in RAM and loses it on every restart. This CLI submits graphs, reads honest timings from the server's own execution timestamps, and keeps runs, graphs, node schemas and outputs in local SQLite so comparisons survive the restarts that tuning requires. It attaches to an in-flight render rather than starting a second one, and it answers questions the server cannot: why a model is invisible, what a loader will actually accept, and which configuration produced a given file.

When to Use This CLI

Reach for this when driving a local ComfyUI server programmatically: submitting graphs, comparing configurations, or tracing a rendered file back to what produced it. It is the right tool when a comparison must survive a server restart, when a model appears invisible to a loader, or when a duration needs to be trustworthy rather than scraped from a log. It is not an installer or a node manager — use comfy-cli for installing ComfyUI and its custom nodes.

Read the full file on GitHub · 569 lines

Files

What ships with it

60 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 569 lines · 81 tokens per session scan A ccf81510bb59

Subscribe to this mod's changes

pp-comfyui is a skill published in the GitHub repository dmmdea/offload-harness (2 stars, last pushed 3d ago), licensed Apache-2.0. It adds 81 tokens to every session and 7,236 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Gemma claude-code cli llamacpp local-llm

python-repl

Interactive Python REPL automation with common helpers and best practices.

gptme/gptme · 15 tokens

implementing-llms-litgpt

Implements and trains LLMs using Lightning AI's LitGPT with 20+ pretrained architectures (Llama, Gemma, Phi, Qwen, Mistral). Use when need clean model implementations, educational understanding of architectures, or production fine-tuning with LoRA/QLoRA. Single-file implementations, no abstraction layers.

Orchestra-Research/AI-Research-SKILLs · 77 tokens

pp-comfyui