agentleFS
JerryLiu369JerryLiu369/awesome-quant-skills/skills/china-markets/ssquant-futures/SKILL.md

ssquant-futures

Skill5 starsChanged 3 days agoMITView on GitHub

Use this compatibility skill only when the runtime discovers capabilities through SKILL.md. The public QUANTSKILLS package is quantskills/agent-ssquant, and the main execution contract is the repository root AGENTS.md.

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 4 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈4

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use this compatibility skill only when the runtime discovers capabilities through SKILL.md.

    line 31
  2. Distinctive:

    Run the doctor when project state is unclear:

    line 37
  3. Very specific:

    Do not call low-level CTP modules, generated binary adapters, DataSource, BacktestCore, or account internals from user strategies.

    line 60
  4. Very specific:

    Do not place data-server credentials, SIMNOW passwords, real CTP passwords, app_id, auth_code, or raw account dictionaries in strategies, …

    line 61
  5. Distinctive:

    Do not claim a strategy, data service, report, or live process works without command output, report files, logs, browser/API evidence, or …

    line 62
  6. Distinctive:

    Do not migrate a profitable backtest to SIMNOW or real trading automatically.

    line 63
  7. Very specific:

    Use data_sources=[...] for multi-symbol or multi-period work.

    line 68
  8. Distinctive:

    Use real backtests, real local data inspection, or real API probes when the user asks for proof.

    line 82
  9. Very specific:

    Read ../../AGENTS.md before running multi-step SSQuant coding, migration, publishing, or live-operation tasks.

    line 86

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md
<!-- Imported from: https://github.com/quantskills/agent-ssquant Source Path: skills/ssquant/SKILL.md Curated for awesome-quant-skills -->

name: ssquant-agent description: Compatibility skill entry for the SSQuant Agent futures workflow. Use when an agent runtime only discovers local capabilities through SKILL.md but the actual public QUANTSKILLS package is quantskills/agent-ssquant with AGENTS.md as the main contract. quantSkills: organization: https://github.com/quantskills repository: quantskills/agent-ssquant repository_url: https://github.com/quantskills/agent-ssquant project_type: skill collection: ssquant license: GPL-3.0 category: tooling tags: [ssquant, agent, futures, ctp, backtest, data-server, reporting] platforms: [claude-code, codex, openclaw, cursor, workbuddy] language: zh-en status: stable validation_level: runnable maintainer_type: community requires: [] summary_zh: "\u672c\u5730\u517c\u5bb9\u5165\u53e3\uff1a\u5c06\u0043\u006f\u0064\u0065\u0078\u7b49\u0053\u004b\u0049\u004c\u004c\u52a0\u8f7d\u5668\u8def\u7531\u5230\u0053\u0053\u0051\u0075\u0061\u006e\u0074\u0020\u0041\u0067\u0065\u006e\u0074\u3002" summary_en: Local compatibility entry that routes SKILL-based runtimes to the SSQuant Agent.

SSQuant Agent Compatibility Skill

Use this compatibility skill only when the runtime discovers capabilities through SKILL.md. The public QUANTSKILLS package is quantskills/agent-ssquant, and the main execution contract is the repository root AGENTS.md.

First Steps

  1. Locate the active SSQuant project from the user's path, current working directory, editable install, or explicit repo path.
  2. Inspect the local project before claiming behavior. Prefer current source and command output over memory.
  3. Run the doctor when project state is unclear:
$env:PYTHONUTF8='1'
python ../../scripts/ssquant_doctor.py --project <ssquant-root>
  1. Choose exactly one primary lane from the routing table.
  2. Compile, validate, and run real evidence when the user asks whether something works.

Routing Table

User intentLoad these files
StrategyAPI return types, project boundaries, validation gates../../references/core/api-return-types.md, ../../references/core/project-contract.md, ../../references/core/verification.md
Futures strategy generation, backtesting, continuous contracts, costs, rollover cost../../references/futures/futures-workflow.md, ../../references/futures/rollover-cost.md
data_server, cache, SQLite, raw vs adjusted prices, async history tasks../../references/data/raw-and-adjustment.md, ../../references/data/cache-and-missing-data.md, ../../references/data/async-history-tasks.md
SIMNOW, real CTP, account config, live K-line source, auto roll, guarded process checks../../references/live-ctp/live-guardrails.md, ../../references/live-ctp/rollover-and-kline-source.md
Backtest metrics, Chinese HTML/Markdown reports, chart/report bugs../../references/reporting/chinese-report-template.md, ../../references/reporting/report-debugging.md

Non-Negotiable Boundaries

  • Generated strategy code stays inside StrategyAPI, initialize(api), strategy(api), get_config(), UnifiedStrategyRunner, and RunMode.
  • Do not call low-level CTP modules, generated binary adapters, DataSource, BacktestCore, or account internals from user strategies.
  • Do not place data-server credentials, SIMNOW passwords, real CTP passwords, app_id, auth_code, or raw account dictionaries in strategies, reports, screenshots, commits, or chat summaries.
  • Do not claim a strategy, data service, report, or live process works without command output, report files, logs, browser/API evidence, or another concrete artifact.
  • Do not migrate a profitable backtest to SIMNOW or real trading automatically. Runtime migration is a separate gated task.

Shared SSQuant Facts

  • A data source is symbol + period + adjust_type + range/limit.
  • Use data_sources=[...] for multi-symbol or multi-period work. Do not teach comma-separated symbol="rb888,hc888" unless the local project proves it.
  • data_server may return raw bars. SSQuant can apply local adjustment according to adjust_type.
  • Strategy continuity may use adjusted prices, while execution/accounting should use raw prices when the framework supports that split.
  • In current SSQuant forks, api.get_price() is a scalar current strategy price; api.get_close(), api.get_open(), api.get_high(), api.get_low(), and api.get_volume() return pandas Series.

Validation Commands

$env:PYTHONUTF8='1'
python ../../scripts/ssquant_doctor.py --project <ssquant-root>
python -m py_compile <strategy.py>
python ../../scripts/validate_strategy.py <strategy.py>

Use real backtests, real local data inspection, or real API probes when the user asks for proof.

Agent Contract

Read ../../AGENTS.md before running multi-step SSQuant coding, migration, publishing, or live-operation tasks. It defines the agent execution contract, evidence requirements, and live-trading gates.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.