← Back to Skills Marketplace
Pyfolio Performance
by
Tang Weigang
· GitHub ↗
· v0.3.2
· MIT-0
105
Downloads
0
Stars
0
Active Installs
3
Versions
Install in OpenClaw
/install pyfolio-performance
Description
基于 pyfolio-reloaded 的投资组合绩效分析：一键生成 tear sheet（夏普、回撤、 年化、换手、个股往返交易、行业归因）。适用于回测后的标准化报告。
Usage Guidance
This skill appears to be a coherent finance backtest/tear-sheet helper, but the packaging is sloppy and some runtime actions are underspecified. Before installing or invoking it: 1) Confirm you trust the skill source (homepage is missing). 2) Expect the agent may run Python commands and install packages (it references pip/zvt/uv) and will read/write a local ZVT_HOME — run it in a sandbox or disposable environment. 3) Verify the missing LICENSE.txt and any dependency list (ask the author for a requirements list or a proper install spec). 4) Do not provide unrelated credentials; if you plan to use paid data providers (joinquant/qmt), supply their keys only after verifying why and where they are used. 5) If you need higher assurance, request a version with an explicit install manifest (requirements.txt or lockfile) and a source/homepage you can audit.
Capability Analysis
Type: OpenClaw Skill
Name: pyfolio-performance
Version: 0.3.2
The pyfolio-performance skill bundle is a highly structured financial analysis tool designed for the OpenClaw/Doramagic ecosystem. It provides comprehensive tear sheet generation and backtesting validation using the pyfolio-reloaded and zvt frameworks. The bundle includes extensive 'semantic locks' and 'domain constraints' (found in SKILL.md and references/seed.yaml) specifically designed to prevent common quantitative finance errors such as look-ahead bias, survivorship bias, and improper transaction cost modeling. The installation recipes and execution protocols are standard for the stated purpose, and no evidence of malicious intent, data exfiltration, or harmful prompt injection was found.
Capability Tags
Capability Assessment
Purpose & Capability
Name/description, use cases and included reference content are consistent with a pyfolio-style performance/tear-sheet tool for backtest results. However SKILL.md declares 'Requires Python 3.12+ with uv package manager' while the registry metadata lists no required binaries or env vars — a packaging inconsistency. The SKILL.md also references a LICENSE.txt that is not present in the manifest.
Instruction Scope
SKILL.md contains runtime directives and preconditions that instruct the agent to run local Python checks and potentially install packages (e.g., 'python3 -c ...', 'pip install zvt'). It also mandates re-reading seed.yaml at execution time and references local data dirs (ZVT_HOME). These instructions are relevant to the skill's purpose but give the agent discretion to execute package installs and write/read local files; the skill does not declare an explicit install recipe or list of dependencies, so the exact runtime actions are under-specified.
Install Mechanism
There is no declared install spec (instruction-only), which lowers supply-chain risk. But SKILL.md implicitly expects Python 3.12+ and the 'uv' package manager and suggests using pip to install 'zvt' if checks fail — so runtime will likely perform package installs without a reviewed install manifest. That gap is a minor risk (unreviewed runtime installs).
Credentials
The skill declares no required environment variables or credentials, which matches that it is an offline analysis/reporting tool. Nevertheless, instructions reference ZVT_HOME and recommend recorders/data providers (eastmoney, joinquant, baostock, akshare, qmt). Use of some data providers (e.g., joinquant/qmt) may require external accounts/credentials in practice, but none are requested up-front — this is under-specified and could prompt the agent to ask for or attempt to use credentials at runtime.
Persistence & Privilege
always is false and disable-model-invocation is false (normal). The skill does not request to be always-enabled nor attempts to modify other skills. It does instruct writing/reading local data directories (ZVT_HOME), which is expected for a backtesting/reporting tool.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install pyfolio-performance - After installation, invoke the skill by name or use
/pyfolio-performance - Provide required inputs per the skill's parameter spec and get structured output
Version History
v0.3.2
v0.3.2: inject bilingual metadata per naming spec. H1 now shows Pyfolio 业绩分析 + slug; tagline and description replaced with CTO-authored copy (fixes tagline pollution for non-ZVT skills).
v0.3.1
Remove install.sh — knowledge-only bundle. Host AI consumes directly from URL; no user-side installation needed. Fixes ClawHub suspicious flag.
v0.3.0
Doramagic crystal portfolio v0.3.0. Full 5-layer bp-009 standard. github.com/tangweigang-jpg/doramagic-skills
Metadata
Frequently Asked Questions
What is Pyfolio Performance?
基于 pyfolio-reloaded 的投资组合绩效分析：一键生成 tear sheet（夏普、回撤、 年化、换手、个股往返交易、行业归因）。适用于回测后的标准化报告。 It is an AI Agent Skill for Claude Code / OpenClaw, with 105 downloads so far.
How do I install Pyfolio Performance?
Run "/install pyfolio-performance" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Pyfolio Performance free?
Yes, Pyfolio Performance is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Pyfolio Performance support?
Pyfolio Performance is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Pyfolio Performance?
It is built and maintained by Tang Weigang (@tangweigang-jpg); the current version is v0.3.2.
More Skills