A skill for creating image batches while changing selected generation settings across a range or list of values. It supports settings such as guidance scale, sampling steps, sampler, scheduler, seed, denoising strength, width, and height.

Comparing image outputs across settings, testing integer or stepped ranges, trying several sampler or scheduler names, and generating multiple random seeds.

It removes the need to create each parameter combination manually when comparing image-generation results. The supported range formats make repeated experiments easier to define.

batch — Command by sandyup

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 commands/sandyup/comfyui-mcp/batch

Clone the repo

git clone --depth 1 https://github.com/sandyup/comfyui-mcp

Per session 6 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,050 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin 89% copy Near-identical to another mod 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	$0.00006	$0.01050
Opus 5	$0.00003	$0.00525
Sonnet 5	$0.00001	$0.00210
Haiku 4.5	$0.00001	$0.00105

Measured yesterday against content hash 2bb8ed2d5c01, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

batch 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 yesterday.

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.

Origin

This is a copy

89% identical to batch — 36 lines differ, which has more behind it and is treated as the original. This page carries a canonical link to it rather than competing with it.

plugin/commands/batch.md · 91 lines

How it starts

The opening of the file, as written. The whole thing — 91 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/comfy-batch — Parameter Sweep Generation

The user wants to generate multiple images while sweeping across different parameter values to compare results.

Instructions

Parse the arguments. The argument is: $ARGUMENTS

If no argument was provided, ask the user for a prompt and which parameters to sweep.

Extract:
- Prompt text: everything that isn't a parameter range specifier
- Parameter ranges: identified by param_name:values syntax
Parse parameter range syntax. Supported formats:
- param:min-max — integer range with step 1 (e.g., cfg:5-10 produces 5, 6, 7, 8, 9, 10)
- param:min-max:step — range with explicit step (e.g., cfg:4-12:2 produces 4, 6, 8, 10, 12)
- param:val1,val2,val3 — explicit list (e.g., sampler:euler,dpmpp_2m,dpmpp_sde)
- seed:N — special: generate N different random seeds (e.g., seed:4 produces 4 random seeds)
Supported sweep parameters:
- cfg — CFG scale (float)
- steps — sampling steps (integer)
- sampler or sampler_name — sampler algorithm name
- scheduler — scheduler name
- seed — random seed count or explicit seeds
- denoise — denoising strength (float, 0.0-1.0)
- width — image width in pixels
- height — image height in pixels
Calculate total combinations. Multiply the count of values for each swept parameter. If the total exceeds 20, warn the user:
- Show the total count and estimated time
- Ask for confirmation before proceeding
- Suggest reducing ranges if the count is very high
Check available models. Call list_local_models with model_type: "checkpoints" to find a checkpoint. If none are available, follow the model acquisition steps from the gen command.
Enqueue all combinations. For each parameter combination:
- Call create_workflow with template "txt2img" and the current parameter set including positive_prompt
- Call enqueue_workflow with the created workflow
- Collect the returned prompt_id

Read the full file on GitHub · 91 lines

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.

yesterday First seen · 91 lines · 6 tokens per session scan A 2bb8ed2d5c01

Subscribe to this mod's changes

batch is a command published in the GitHub repository sandyup/comfyui-mcp (1 stars, last pushed 1mo ago), licensed MIT. It adds 6 tokens to every session and 1,050 once invoked, about $0.0000 per session on Opus 5. A static security scan graded it A with 0 findings. It is 89% identical to batch, differing in 36 lines, and is treated as a copy.

Testing

batch

Getting it into your agent

What it costs to keep this loaded

Grade A, and why

Nothing flagged

This is a copy

How it starts

/comfy-batch — Parameter Sweep Generation

Instructions

What this file has done since we first saw it

Other commands, from other repositories

checklist

clarify

specify

analyze

converge

implement