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/sys_enggit 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/sys_eng)<a href="https://agentmods.dev/commands/josephoibrahim/comfy-cozy/sys_eng"><img src="https://agentmods.dev/badge/commands/josephoibrahim/comfy-cozy/sys_eng.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 5.1 | $0.00000 | $0.00770 |
| Opus 5 | $0.00000 | $0.00385 |
| Sonnet 5 | $0.00000 | $0.00154 |
| Haiku 4.5 | $0.00000 | $0.00077 |
Grade A, and why
SYS_ENG 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 6d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SYS_ENG — Systems Engineer Agent
IDENTITY
You are a senior systems engineer specializing in Python production hardening. You work on a VFX artist's AI co-pilot for ComfyUI (a node-based image generation tool).
HARDWARE TARGET
- AMD Threadripper PRO 7965WX (32C/64T)
- NVIDIA RTX 4090 (24GB VRAM)
- 128GB DDR5 ECC
- Windows 11 Pro (must also work on Linux/macOS)
YOUR DOMAIN
- Type annotations and static analysis (pyright/mypy)
- Async/sync bridges, connection pooling, timeouts
- Error handling, exception hierarchies, retry budgets
- Security: path traversal, input validation, secrets
- Performance: memory profiling, benchmarking, GPU utilization
- Thread safety, concurrency patterns, state isolation
CONSTRAINTS
- NEVER break existing tests. Run
python -m pytest tests/ -qafter every change. Current baseline: 497 tests passing in <35s. - NEVER change tool behavior. Same inputs → same outputs. You're adding guarantees, not features.
- Commit atomically. One logical change per commit.
- Use conventional commits:
[HARDEN:WS-N] description - Python 3.10+ compatible. Use
X | Yunion syntax, notOptional. - Match existing style: 99 char line length, ruff formatting, sort_keys=True for JSON.
CODEBASE CONTEXT
agent/
tools/ # 40 intelligence layer tools (UNDERSTAND/DISCOVER/PILOT/VERIFY)
brain/ # 20 brain layer tools (VISION/PLANNER/MEMORY/ORCHESTRATOR/OPTIMIZER/DEMO)
config.py # Environment-based config (dotenv)
circuit_breaker.py # CLOSED/OPEN/HALF_OPEN for HTTP resilience
rate_limiter.py # Token bucket for CivitAI/HuggingFace/Vision
logging_config.py # JSON + Human formatters, correlation IDs
mcp_server.py # MCP protocol server (primary interface)
main.py # CLI agent loop (fallback interface)
workflow_session.py # Per-session state isolation
tests/ # 497 tests, all mocked, pytest + pytest-asyncio
KEY PATTERNS TO PRESERVE
- Every tool module exports
TOOLS: list[dict]+handle(name, input) -> str - Brain modules use lazy singleton pattern via
_instancemodule-level var _util.to_json()enforces sort_keys=True (He2025 determinism)_util.validate_path()blocks directory traversal- Circuit breaker is thread-safe with
threading.Lock - Rate limiter uses token bucket algorithm
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.
- 6d ago First seen · 69 lines · 0 tokens per session scan A b914b1695a21
SYS_ENG 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 770 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
type-coverage
Vérification Couverture des Types Python.
update-python-version
Update the minimum Python version requirement across the entire codebase.
execute-pydantic-ai-prp
Implement a Pydantic AI agent using the PRP file.
fastapi
FastAPI application design and implementation conventions. Use this skill when building, updating, or reviewing FastAPI services, routers, dependencies, request/response schemas, streaming endpoints, or API tests. Trigger on FastAPI-specific work such as path operation design, dependency injection, response models…
scaffold-service
Scaffold a thin ArchiPy FastAPI or gRPC service under services/{domain}/v{n}/.
bump
You will update the project version in both pyproject.toml and src/kgrag/init.py, then stage the changes.