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 skills/policyengine/policyengine-claude/policyengine-uknpx skills add PolicyEngine/policyengine-claude --skill policyengine-ukgit clone --depth 1 https://github.com/PolicyEngine/policyengine-claudeWhat 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 | $0.00241 | $0.03607 |
| Opus 5 | $0.00120 | $0.01803 |
| Sonnet 5 | $0.00048 | $0.00721 |
| Haiku 4.5 | $0.00024 | $0.00361 |
Grade A, and why
policyengine-uk 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 3d 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 — 304 lines — stays where its author put it; the contents beside it link to each section on GitHub.
PolicyEngine UK
UK-specific facts for the PolicyEngine Python stack. Read the policyengine skill first for
the canonical import policyengine as pe interface, the population flow
(ensure_datasets / Simulation / economic_impact_analysis), the MicroSeries discipline,
and dataset handling — this skill only carries what is different about the UK.
Verified against policyengine 4.21.0 + policyengine-uk 2.89.2 + policyengine-core 3.30.0
(2026-07). UK examples are not in the [us]-only CI verify set, so they carry no
<!-- verify --> marker; each was run manually in a policyengine[uk] environment.
Setup
The UK model is an extra — bare policyengine installs no country model:
uv pip install "policyengine[uk]" # UK model + certified UK data bundle metadata
Household calculations need no data download. Population runs need the UK dataset, which is a private Hugging Face repo — see "Population data" below.
Entity model — and the benunit trap
UK entities are person → benunit → household. There is no tax_unit and no
spm_unit / marital_unit / family — those are US entities. The single most common UK
mistake is reaching for result.tax_unit out of US habit; on a UK result it raises
AttributeError (verified: a UK result has person, benunit, household and no tax_unit).
| Concept | US entity | UK entity |
|---|---|---|
| Means-tested assessment unit | tax_unit (+ spm_unit) |
benunit (benefit unit) |
| Statistical unit for poverty | spm_unit / household |
household |
| Individual | person |
person |
A benefit unit is a single adult or a couple plus their dependent children — the DWP assessment unit for means-tested benefits (Universal Credit, Pension Credit, etc.). One household can contain several benunits (e.g. adult children living with parents). Read means-tested benefits at the benunit level, income tax and NI at the person level, and net income and poverty at the household level.
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.
- 3d ago First seen · 304 lines · 241 tokens per session scan A cfeb7c4b7d29
policyengine-uk is a skill published in the GitHub repository PolicyEngine/policyengine-claude (31 stars, last pushed 8d ago), licensed MIT. It adds 241 tokens to every session and 3,607 once invoked, about $0.0012 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 2026-08-30.
Other skills, from other repositories
tqsdk-trading-and-data
Explain, implement, or debug TqSdk Python workflows for waitupdate or ischanging update loops, market data retrieval, historical download, account type selection, funds or positions or orders or trades, field meanings, order placement or cancellation, target-position tools, TqScenario margin trials, real-account…
miniqmt-to-bigqmt
将基于 miniQMT 的外部 xtquant 库（xttrader/xtdata）的 Python 量化策略，转换为大QMT内置Python（ContextInfo/passorder 体系）可实盘运行的策略。当用户要求转换 miniQMT 策略、迁移 xtquant 代码到大QMT、或提到"内置Python/XTData/passorder 改写"时使用。包含可行性评估、API 映射、py3.6+GBK 约束校验、部署与实盘验证全流程；不可转换场景给出文件桥等替代方案。.
python-binance
Help developers use the python-binance library for trading on Binance. Use when code imports binance, references Client/AsyncClient, or asks about Binance API trading, market data, websockets, or account management.
shioaji
Use for Shioaji, SJ, SinoPac (永豐金), Taiwan market trading/API tasks, or broad order/trading requests where Shioaji may apply. Covers Python sync/async bindings, shioaji CLI, HTTP API, SSE streaming, dashboard embedding, and JS/TS, Go, C/C++, C#, Rust, Java/Kotlin HTTP clients. Covers TWSE/TPEX/TAIFEX orders, quotes…
yfinance
Load proactively whenever the user works with yfinance or Yahoo Finance data — pulling price history, fetching financials or analyst data, screening stocks / funds / ETFs, streaming real-time quotes, or inspecting sector, industry, or fund rollups. Do not wait to be asked; apply this skill automatically whenever the…
tdx-quant-docs
当用户询问 TdxQuant/tqcenter API 用法、通达信 Python 量化策略开发、或需要查阅 TdxQuant 文档时，务必使用此技能。即使用户未显式提及…