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.
npx agentmods add commands/josephoibrahim/comfy-cozy/quickstartgit clone --depth 1 https://github.com/JosephOIbrahim/Comfy-CozyWrote 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.
[](https://agentmods.dev/commands/josephoibrahim/comfy-cozy/quickstart)<a href="https://agentmods.dev/commands/josephoibrahim/comfy-cozy/quickstart"><img src="https://agentmods.dev/badge/commands/josephoibrahim/comfy-cozy/quickstart.svg" alt="Measured on agentmods" height="20"></a>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 | $0.00000 | $0.01229 |
| Opus 5 | $0.00000 | $0.00615 |
| Sonnet 5 | $0.00000 | $0.00246 |
| Haiku 4.5 | $0.00000 | $0.00123 |
Grade A, and why
QUICKSTART 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 4d 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.
How it starts
The opening of the file, as written. The whole thing — 114 lines — stays where its author put it; the contents beside it link to each section on GitHub.
⚠️ INTERNAL MoE-team reference — STALE. Not user-facing. References
bootstrap_hardening.py/.claude/agents/orchestrate.py(removed) and the pre-4,744-test "497+ pass" count. Kept for archaeology; do not follow its commands. User quickstart lives at repo-rootQUICKSTART.md.
MoE AGENT TEAM — QUICK REFERENCE
INSTALL (run from comfyui-agent/)
mkdir -p .claude/agents
# Copy all files from this delivery into .claude/agents/
# Copy PRODUCTION_HARDEN.md to project root
python bootstrap_hardening.py
RUN AGENTS
Option A: Orchestrator Script (recommended)
# Preview what will happen
python .claude/agents/orchestrate.py --phase 1 --dry-run
# Execute Phase 1 (foundation — type safety, error handling, deprecation fixes)
python .claude/agents/orchestrate.py --phase 1
# Execute Phase 2 (core — async, state, security) — can run in parallel
python .claude/agents/orchestrate.py --phase 2
# Execute single workstream
python .claude/agents/orchestrate.py --workstream WS-6
# Check status
python .claude/agents/orchestrate.py --status
# Run everything
python .claude/agents/orchestrate.py --all
Option B: Direct Claude Code Invocation
# Systems Engineer — Type Safety
claude --model claude-sonnet-4-6-20250929 \
--system-prompt "$(cat .claude/agents/SYS_ENG.md)" \
--prompt "Execute workstream WS-1: Type Safety & Static Analysis. Branch: harden/sys-types. See PRODUCTION_HARDEN.md for full task list. Commit after each task. Run tests after each commit. GO."
# ComfyUI Lead — State Management
claude --model claude-sonnet-4-6-20250929 \
--system-prompt "$(cat .claude/agents/COMFY_LEAD.md)" \
--prompt "Execute workstream WS-4: State Management Hardening. Branch: harden/comfy-state. See PRODUCTION_HARDEN.md. GO."
# VFX Supervisor — Documentation
claude --model claude-sonnet-4-6-20250929 \
--system-prompt "$(cat .claude/agents/VFX_SUPER.md)" \
--prompt "Execute workstream WS-9: Documentation & UX. Branch: harden/vfx-docs. See PRODUCTION_HARDEN.md. GO."
# Nuke Compositor — Platform Config
claude --model claude-sonnet-4-6-20250929 \
--system-prompt "$(cat .claude/agents/NUKE_COMP.md)" \
--prompt "Execute workstream WS-10: Platform & Config Hardening. Branch: harden/pipeline-config. See PRODUCTION_HARDEN.md. GO."
# Producer — CI/CD
claude --model claude-sonnet-4-6-20250929 \
--system-prompt "$(cat .claude/agents/PRODUCER.md)" \
--prompt "Execute workstream WS-8: CI/CD & Packaging. Branch: harden/ops-cicd. See PRODUCTION_HARDEN.md. GO."
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.
- 4d ago First seen · 114 lines · 0 tokens per session scan A d5dcb3270256
QUICKSTART is a command published in the GitHub repository JosephOIbrahim/Comfy-Cozy (24 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,229 tokens. 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.
Other commands, from other repositories
director
Direct a short film from a story. Generates scenes, frames, and video clips.
gen
Generate an image with ComfyUI from a text prompt.
batch
Parameter sweep generation across multiple values.
compare
Diff two ComfyUI workflows to see what changed.
install
Install a ComfyUI custom node pack.
recipe
Run a multi-step generation recipe (portrait, hires-fix, style-transfer, etc.).