Qc Backtest Master
v1.0.0QuantConnect automated backtest pipeline — submit local strategies, compile, execute, monitor with early-stop, and download results in one command.
⭐ 0· 144·0 current·0 all-time
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (QuantConnect backtest pipeline) match the code and required environment variables (QC_USER_ID, QC_API_TOKEN, QC_PROJECT_ID). Required binary (python3) and the API host (https://www.quantconnect.com/api/v2) are appropriate for the stated purpose. Deleting/overwriting project files and calling compile/backtest endpoints are consistent with a tool that automates submission and monitoring.
Instruction Scope
SKILL.md and scripts instruct uploading a local main strategy and auxiliary data files, compiling, monitoring, optionally deleting backtests, and downloading results — all within the QuantConnect API. A couple of scope notes: the code will upload any .json/.txt/.csv in the strategy directory (could unintentionally include sensitive data files), and monitor_backtest supports automatic deletion of backtests on early-stop (documented). These behaviors are coherent with the skill's purpose but are important safety/usability points for users to understand.
Install Mechanism
No install spec is provided (instruction-only skill plus shipped Python scripts). requirements.txt lists only 'requests', which is proportionate. The code runs local Python scripts and uses subprocess to chain steps; no remote arbitrary downloads or obscure install sources are present.
Credentials
Requested environment variables (QC_USER_ID, QC_API_TOKEN, QC_PROJECT_ID) are appropriate and declared; QC_API_TOKEN is marked primary. The config module also attempts to load a .env up to 6 directory levels and falls back to ~/.lean/credentials — reasonable convenience for QuantConnect users but worth noting because it may pick up credentials from nearby .env files unexpectedly. The skill does not request unrelated credentials.
Persistence & Privilege
always:false and default autonomous invocation are appropriate. The skill does not request system-wide changes or modify other skills. It reads local files (strategy and data) and user credential files for operation, which is expected for this capability.
Assessment
This skill appears to do exactly what it claims: submit your local strategy to QuantConnect, compile, run/monitor it (with optional early-stop that permanently deletes the backtest), and fetch results. Before using it: 1) Only provide QC_USER_ID/QC_API_TOKEN/QC_PROJECT_ID for an account you trust; the token is used to authenticate to QuantConnect. 2) Run the skill from a directory that does not contain sensitive .json/.txt/.csv files or an unrelated .env you don't want uploaded — the submit step uploads all such files in the strategy directory. 3) Be aware early-stop will permanently delete the backtest (documented). 4) The code will also look for credentials in a parent .env (up to 6 levels) and in ~/.lean/credentials — remove or inspect those if you don't want them picked up automatically. 5) If you have any doubt, review the included Python files locally (they call only QuantConnect API endpoints) and run verify_connection.py first to confirm the connection behaves as expected.Like a lobster shell, security has layers — review code before you run it.
latestvk97b5kpp2e400b4me65atkx27s833w95
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🚀 Clawdis
Binspython3
EnvQC_USER_ID, QC_API_TOKEN, QC_PROJECT_ID
Primary envQC_API_TOKEN