← Back to Skills
Quant · No market data required Strategy rule specification Make a trading idea explicit and find the gaps before execution or testing.
See what you will get
Turn an idea into rules you can check. Synthetic teaching sample · not a real company / fund Turn an idea into rules you can check. Condition Teaching rule Origin / status Assets and weights ETF_A 60% / ETF_B 40% User idea Rebalance / benchmark Month end / equal-weight buy and hold User idea Signal → execution After close → next valuation close Proposed; confirm first Friction / cash return 10 bps per side / 0 Proposed; confirm first Initial value / currency 10,000 / synthetic SYNTH units Proposed; confirm first
ETF_A and ETF_B are teaching identifiers. Set confirmed only after explicit acceptance, recording time and rule digest. Otherwise return needs-input.
Keep the records in your workspace Rule card, cost assumptions and data requirements
strategy-card.mdstrategy.jsondata-requirements.mdconfirmation.md
Each run writes to a new reports/{run-id}/, preserving inputs and previous reports. Status distinguishes completed, needs-input and failed.
Start with this request Use the strategy specification Skill to turn my idea into testable rules. Separate my decisions, your proposed defaults and unresolved questions. Specify signal and execution timing. Do not run a backtest yet. First ask your host to inspect the method version, inputs, output location and dependencies in read-only mode. Downloading, discovering a Skill and completing its sample are separate steps.
Its limits This step does not run a backtest. Suggested defaults require actual confirmation, and changed rules need confirmation again. Unsupported strategies receive a draft and limits.
No network retrieval, account connection or result upload by default. Model handling depends on your host and service settings; this is not a promise of fully offline processing.
Use it in your AI Choose a tool you already use. Codex CLI is selected initially; you can switch at any time.
Before running: all these methods require Python 3.12. Bundled calculation tools use only its standard library. If it is missing, stop and follow the setup guide. Bring your own model account or credits.
AI host
JavaScript is disabled; complete instructions for both hosts appear below.
Use this Skill → Close usage steps Codex CLI Sample task acceptance in this host is pending. These preparation steps do not imply a pass.
Get and inspect the package. Review the version, license, guide and reproducible examples. A build exists, but host acceptance is incomplete. No recommended download yet.
Install this package only. Copy the complete folder into your personal skills directory. SKILL.md must be directly inside the skill folder.%USERPROFILE%\.agents\skills\cotrading-strategy-spec\SKILL.mdStop if the folder already exists. Preserve its version and local edits before choosing an update.Confirm discovery. Use /skills or type $ in your conversation to select a Skill.$cotrading-strategy-specThis invocation belongs in the AI conversation, not PowerShell. If missing, check nesting, name and conflicts, then reopen the session if needed.
Run the sample, then use your materials. Copy examples to a separate research workspace. First inspect dependencies, inputs and output location in read-only mode. Confirm new dependencies or external access as needed, then check evidence or arithmetic. Installation & first-run guide → Version & integrity details File cotrading-strategy-spec-0.1.0-codex.zip SHA-256 c4fbf109c62c8dbb0f81d4137e6b60b80b696264e99b7686238959e570d3788c Size 67043 bytes Host acceptance evidence /skill-packages/verification.json Claude Code Sample task acceptance in this host is pending. These preparation steps do not imply a pass.
Get and inspect the package. Review the version, license, guide and reproducible examples. A build exists, but host acceptance is incomplete. No recommended download yet.
Install this package only. Copy the complete folder into your personal skills directory. SKILL.md must be directly inside the skill folder.%USERPROFILE%\.claude\skills\cotrading-strategy-spec\SKILL.mdStop if the folder already exists. Preserve its version and local edits before choosing an update.Confirm discovery. Type / in your conversation and choose the Skill./cotrading-strategy-specThis invocation belongs in the AI conversation, not PowerShell. If missing, check nesting, name and conflicts, then reopen the session if needed.
Run the sample, then use your materials. Copy examples to a separate research workspace. First inspect dependencies, inputs and output location in read-only mode. Confirm new dependencies or external access as needed, then check evidence or arithmetic. Installation & first-run guide → Version & integrity details File cotrading-strategy-spec-0.1.0-claude.zip SHA-256 193c413d23f666f60034837cdd6d552eab6779b55ffc5cbc5dd0ac9635b94870 Size 67048 bytes Host acceptance evidence /skill-packages/verification.json
Method package Free
Site account Not required
Paid data Not required
AI usage Your account / credits 