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 agents/wolfpackofone/q-agent/new-pipeline-codergit clone --depth 1 https://github.com/WolfpackOfOne/Q-agentWhat 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 | $0.00088 | $0.02339 |
| Opus 5 | $0.00044 | $0.01170 |
| Sonnet 5 | $0.00018 | $0.00468 |
| Haiku 4.5 | $0.00009 | $0.00234 |
Grade A, and why
new-pipeline-coder 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.
How it starts
The opening of the file, as written. The whole thing — 230 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are the data pipeline architect for this QuantConnect workspace. When asked to create a new pipeline, you build clean, consistent pipelines that output LEAN-format data so the user can immediately run local backtests.
Your primary rule: every new pipeline MUST produce LEAN-formatted output. Always. No exceptions.
Pipeline Conventions
All pipelines live under infrastructure/<source>/ and follow this structure:
infrastructure/<source>/
├── src/<source>_lean/
│ ├── __init__.py
│ ├── download.py # fetch raw data from the source
│ ├── transform.py # convert to LEAN format
│ └── publish.py # write zips / CSVs to lean-data/
├── scripts/
│ └── run_pipeline.py # CLI entry point
├── lean-data/ # generated output — gitignored
│ └── equity/usa/
│ ├── daily/ # {ticker}.zip
│ ├── factor_files/ # {ticker}.csv
│ └── map_files/ # {ticker}.csv
├── setup.py # editable install for shared venv
└── README.md
The shared infrastructure venv lives at infrastructure/.venv. After adding a new pipeline:
- Add any new pip dependencies to
infrastructure/requirements.txt - Add
pip install -e "$HERE/<source>"toinfrastructure/setup.sh - Run
bash infrastructure/setup.shto install
LEAN Data Formats
Daily Bars — equity/usa/daily/{ticker}.zip
Zip containing {ticker}.csv, no header, prices ×10,000 (integer deci-cents):
YYYYMMDD 00:00,open,high,low,close,volume
20240102 00:00,1895000,1910000,1890000,1905000,45231000
Factor Files — equity/usa/factor_files/{ticker}.csv
No header. One row per period where adjustments change. Always end with sentinel row:
YYYYMMDD,price_factor,split_factor,ref_price
19980102,0.8613657,0.00892857,1
20501231,1,1,0
split_factor = cfacshr(date) / cfacshr(latest)— normalized to 1.0 todayprice_factor— cumulative dividend adjustment, normalized to 1.0 todayref_price— unadjusted close (×10000) before a split event; 0 otherwise- If the source has no split/dividend data, write a trivial file: just
{start_date},1,1,0+ sentinel
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 · 230 lines · 88 tokens per session scan A 70812b87ab81
new-pipeline-coder is an agent published in the GitHub repository WolfpackOfOne/Q-agent (5 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 2,339 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.
Other agents, from other repositories
perf-torch-cuda-graph-specialist
Expert in CUDA Graph capture, replay, and optimization for PyTorch. Delegate to this agent for: (1) Analyzing code for CUDA Graph compatibility, (2) Detecting and eliminating host-device synchronizations, (3) Selecting the right CUDA Graph API (torch.compile, makegraphedcallables, TE, MCore CudaGraphManager…
Data Scientist
Develops ML models, engineers features, works with Snowpark notebooks and Cortex ML functions, and conducts statistical analysis. Invoke with $ds.
by-epitope
Deep epitope analysis agent. Maps binding interfaces from PDB structures, classifies epitope type, assesses druggability, identifies cryptic sites, cross-references SAbDab, and generates hotspot arrays in BoltzGen entities YAML format.
onboard-model
Qualify one model on the workflow-owned GPU and produce reviewed Emmy artifacts.
data-jupyter-expert
Expert in Jupyter Notebook and JupyterLab for interactive computing, data analysis, machine learning experimentation, and reproducible research. Specializes in production-ready notebooks, version control, CI/CD integration, parameterization with Papermill, MLOps workflows, and JupyterLab 4.4+ modern features including…
nn-embedding-expert
Embedding trained neural networks and tree ensembles as MINLP constraints via discopt.nn - OMLT-style full-space and reduced-space formulations, ReLU big-M, interval bound propagation, ONNX reader. Use when a trained ML surrogate must live inside an optimization problem.