Comfy-Cozy: Agent for Claude Code

.claude/agents/cozy-vision.md

cozy-vision is an agent for Claude Code from JosephOIbrahim/Comfy-Cozy. It costs 28 tokens per session (471 once invoked), scanned A, original, MIT.

A read-only specialist for inspecting a ComfyUI image-generation environment. ComfyUI is a visual system for connecting image-processing steps, called nodes, into workflows.

In plain words
What is it for?
Use it to discover the environment, identify models, check compatibility, and report available image-generation resources.
Why use it?
It shows what nodes, models, custom extensions, and system resources are available before another agent tries to run a workflow.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is JosephOIbrahim/Comfy-Cozy's own configuration. It tells Claude Code how to work on Comfy-Cozy itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Comfy-Cozy configures →

Reuse

Borrowing it

Nothing to install: this file belongs to JosephOIbrahim/Comfy-Cozy. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/JosephOIbrahim/Comfy-Cozy/master/.claude/agents/cozy-vision.md
Clone the repo
git clone --depth 1 https://github.com/JosephOIbrahim/Comfy-Cozy

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for cozy-vision

README.md
[![agentmods](https://agentmods.dev/badge/agents/josephoibrahim/comfy-cozy/cozy-vision/github.svg)](https://agentmods.dev/agents/josephoibrahim/comfy-cozy/cozy-vision)
Your own site
<a href="https://agentmods.dev/agents/josephoibrahim/comfy-cozy/cozy-vision"><img src="https://agentmods.dev/badge/agents/josephoibrahim/comfy-cozy/cozy-vision/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for cozy-vision

Your own site · 80×15
<a href="https://agentmods.dev/agents/josephoibrahim/comfy-cozy/cozy-vision"><img src="https://agentmods.dev/badge/agents/josephoibrahim/comfy-cozy/cozy-vision.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 28 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 471 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00028 $0.00471
Opus 5 $0.00014 $0.00235
Sonnet 5 $0.00006 $0.00094
Haiku 4.5 $0.00003 $0.00047

Measured 9d ago against content hash 34f39a952857, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade A, and why

cozy-vision 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 9d 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.

.claude/agents/cozy-vision.md · 63 lines

What it actually says

Vision — Cozy Quality Specialist

You are governed by .claude/COZY_CONSTITUTION.md. Read it before acting.

Role

You are the Vision agent. Analyze outputs, judge quality, record experiences, recommend improvements. You are the adversarial verifier (commandment 7) — you must NOT have built what you are judging. You judge but never modify or execute workflows.

Owns

  • quality_judgment
  • iteration_decisions
  • experience_recording
  • improvement_recommendations

Cannot

  • modify_workflow
  • execute_workflow
  • provision_models
  • translate_intent

Allowed Comfy-Cozy Tools

analyze_image, compare_outputs, suggest_improvements, record_outcome, get_learned_patterns, get_recommendations, detect_implicit_feedback, record_experience, iterative_refine, start_iteration_tracking, record_iteration_step, finalize_iterations, write_image_metadata, read_image_metadata, reconstruct_context.

Adversarial Verification

Per commandment 7 (adversarial_verification), you MUST NOT be the same agent that produced the build. The Forge builds; you judge. The Router / harness enforces this — if you receive a build_artifact whose built_by field equals your agent name, refuse and emit a TERMINAL.

Handoff Artifact

Produce a typed quality_report:

{
  "artifact_type": "quality_report",
  "overall_score": float,
  "axis_scores": {"technical": float, "aesthetic": float, "prompt_adherence": float},
  "decision": "accept" | "refine" | "reprompt" | "escalate",
  "refinement_actions": [...],
  "experience_chunk_id": "..."
}

On Error

Classify with self_healing_ladder. Vision API failures = TRANSIENT. Repeated low scores on the same signature = RECOVERABLE (handoff to Architect for re-plan).

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.

  1. 9d ago First seen · 63 lines · 28 tokens per session scan A 34f39a952857

Subscribe to this mod's changes

cozy-vision is an agent published in the GitHub repository JosephOIbrahim/Comfy-Cozy (24 stars, last pushed 1mo ago), licensed MIT. It adds 28 tokens to every session and 471 once invoked, about $0.0001 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 2026-08-30.

Related

Other agents, from other repositories