fast-ablation
用于在当前量化仓库中做小切片快实验，参考 vectorbt 的快迭代思路，但优先复用现有脚本和数据切片。用户要快速比较基线与一个变体、先拿真实输出再决定是否上全量时使用。
Skill by thuwindy
Chalk Labs
Need help deploying fast-ablation?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/thuwindy-fast-ablation
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/fast-ablation
What's inside
- Skill name
fast-ablation- Description
- 用于在当前量化仓库中做小切片快实验，参考 vectorbt 的快迭代思路，但优先复用现有脚本和数据切片。用户要快速比较基线与一个变体、先拿真实输出再决定是否上全量时使用。
- License
- MIT
- Instructions
- 42 lines in SKILL.md · view on GitHub
Agent Skills spec issues
- files: referenced but missing: scripts/run_ablation_suite.py, scripts/run_quick_topn_notrade_grid.py, scripts/run_candidate_rerank_validation.py
Files (2)
- SKILL.md
- experiment_matrix.md · references/
SKILL.md (first lines)
# fast-ablation 只做一件事：小切片、快实验、快出真结果。 ## 什么时候用 - 用户要先验证一个新因子/新标签/新权重 - 用户不想一上来跑全量主库 - 用户要对比基线和一个变体 ## 先看什么 1. `references/experiment_matrix.md` 2. 现成脚本： - `scripts/run_ablation_suite.py` - `scripts/run_quick_topn_notrade_grid.py` - `scripts/run_candidate_rerank_validation.py` ## 工作规则 1. 一次只改一个变量 2. 先锁数据切片 3. 先跑基线，再跑变体 4. 不要一次堆多个改动 5. 结果先看排序和净增量，不先看炫目的 accuracy ## 推荐默认切片 - `recent_slice_20240101_20260330_premium_dynamic.csv` ## 推荐输出 | 方案 | 数据区间 | 改动点 | top5 | top10 | 年化 | Sharpe | 回撤 | 结论 | | --- | --- | --- | ---: | ---: | ---: | ---: | ---: | --- | ## 什么时候结束 - 能证明有增量，再上正式验证 - 没增量就停，不继续烧算力
From thuwindy/a-share-quant-codex, MIT.
More from thuwindy/a-share-quant-codex
- 0.476score
qlib-alpha
thuwindy
用于在当前 A 股量化仓库中引入 Qlib 风格 alpha 思路，只抽取对日 K 短线选股有用的因子族，优先做低侵入映射、候选因子清单、落地顺序与最小实验设计。用户要增厚主策略 alpha、把 Qlib Alpha158 或 Alpha360 思路映射到现有因子系统时使用。
SkillCFreshness: CClaude Code +9 - 0.473score
alpha-prototype
thuwindy
用于从 ML4Trading 和 Alpha101 里抽取少量高价值 alpha 原型，在当前 A 股量化仓库中做低侵入试验。用户要找新的 alpha 原型但不想整包接入外部库时使用。
SkillCFreshness: CClaude Code +9 - 0.469score
factor-check
thuwindy
用于对新增 alpha 因子做入库前检验，强调时间对齐、无未来函数、分层命中率、IC、和现有主策略的独立增量。用户要验证一个新因子值不值得接入主分时使用。
SkillCFreshness: CClaude Code +9 - 0.469score
a-share-factor-research
thuwindy
Use when asked to add, modify, or evaluate a stock-selection factor for the A-share quant research system.
SkillCFreshness: CClaude Code +9 - 0.469score
backtest-audit
thuwindy
Use when asked to audit the backtest engine, portfolio construction, transaction costs, or A-share trading constraints.
SkillCFreshness: CClaude Code +9 - 0.469score
research-report
thuwindy
Use when asked to summarize experiment outputs into a research note, course project write-up, or presentation-ready markdown.
SkillCFreshness: CClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Apr 30, 2026 (6 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/thuwindy/a-share-quant-codex
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from thuwindy
- 0.476score
qlib-alpha
thuwindy
用于在当前 A 股量化仓库中引入 Qlib 风格 alpha 思路，只抽取对日 K 短线选股有用的因子族，优先做低侵入映射、候选因子清单、落地顺序与最小实验设计。用户要增厚主策略 alpha、把 Qlib Alpha158 或 Alpha360 思路映射到现有因子系统时使用。
SkillCFreshness: CClaude Code +9 - 0.473score
alpha-prototype
thuwindy
用于从 ML4Trading 和 Alpha101 里抽取少量高价值 alpha 原型，在当前 A 股量化仓库中做低侵入试验。用户要找新的 alpha 原型但不想整包接入外部库时使用。
SkillCFreshness: CClaude Code +9 - 0.469score
factor-check
thuwindy
用于对新增 alpha 因子做入库前检验，强调时间对齐、无未来函数、分层命中率、IC、和现有主策略的独立增量。用户要验证一个新因子值不值得接入主分时使用。
SkillCFreshness: CClaude Code +9 - 0.469score
a-share-factor-research
thuwindy
Use when asked to add, modify, or evaluate a stock-selection factor for the A-share quant research system.
SkillCFreshness: CClaude Code +9 - 0.469score
backtest-audit
thuwindy
Use when asked to audit the backtest engine, portfolio construction, transaction costs, or A-share trading constraints.
SkillCFreshness: CClaude Code +9 - 0.469score
research-report
thuwindy
Use when asked to summarize experiment outputs into a research note, course project write-up, or presentation-ready markdown.
SkillCFreshness: CClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get fast-ablation running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
fast-ablation skill. Agent Mart. https://agentmart.dev/skills/thuwindy-fast-ablation (last updated 2026-10-24)