Borrowing it
Nothing to install: this file belongs to Lambenthan/empiricalwiki. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/Lambenthan/empiricalwiki/main/.claude/skills/refine/SKILL.mdgit clone --depth 1 https://github.com/Lambenthan/empiricalwikiWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/lambenthan/empiricalwiki/refine)<a href="https://agentmods.dev/skills/lambenthan/empiricalwiki/refine"><img src="https://agentmods.dev/badge/skills/lambenthan/empiricalwiki/refine.svg" alt="Measured on agentmods" height="20"></a>What 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 5.1 | $0.00035 | $0.02614 |
| Opus 5 | $0.00017 | $0.01307 |
| Sonnet 5 | $0.00007 | $0.00523 |
| Haiku 4.5 | $0.00003 | $0.00261 |
Grade A, and why
refine 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 7d 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 — 231 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/refine
通用多轮迭代改进循环,适用于任何研究制品(idea、proposal、experiment plan、paper draft)。 每轮调用 /review 获取结构化反馈 → 解析 actionable items → Claude 修复制品 → 更新 wiki 实体 → 重新 /review,直到评分达到目标分数或达到最大轮次。 输出改进历史和最终 review 评分。
Inputs
artifact:要改进的制品,以下之一:- wiki 页面的 slug(从 ideas/experiments/claims/outputs/ 中查找)
- 文件路径(如
wiki/outputs/paper-draft-v1.md)
--max-rounds N(可选,默认 4):最大迭代轮次--target-score N(可选,默认 8):目标 review 评分(1-10),达到后停止--difficulty(可选,默认hard):传递给 /review 的难度级别--focus(可选):传递给 /review 的审查焦点
Outputs
- 改进后的 artifact(wiki 页面或文件,原地更新)
- wiki 实体更新(若 review 发现 claim 需加强或 gap 被识别)
- REFINE_REPORT(输出到终端):
- 每轮的评分变化轨迹
- 累计修复的 issues 列表
- 最终 review 评分和 verdict
- 未解决的 issues(若有)
Wiki Interaction
Reads
wiki/ideas/*.md— 若 artifact 是 ideawiki/experiments/*.md— 若 artifact 是 experiment planwiki/claims/*.md— review 引用的 claimswiki/papers/*.md— review 引用的 paperswiki/outputs/*.md— 若 artifact 是 paper draft 或 outputwiki/graph/context_brief.md— 传递给 /review 的全局上下文wiki/graph/open_questions.md— 检查是否有新 gap 需要记录
Writes
wiki/ideas/{slug}.md— 若 artifact 是 idea,修复 review 发现的问题wiki/experiments/{slug}.md— 若 artifact 是 experiment planwiki/claims/{slug}.md— 若 review 发现 claim 需要更新(confidence 调整、evidence 补充说明)wiki/outputs/*.md— 若 artifact 是 paper draft 或 outputwiki/graph/edges.jsonl— 若修复过程中发现新关系wiki/graph/context_brief.md— 每轮结束后重建(若 wiki 有变更)wiki/graph/open_questions.md— 每轮结束后重建(若 wiki 有变更)wiki/log.md— 追加操作日志
Graph edges created
- 视修复内容而定,可能添加:
supports、addresses_gap、inspired_by等
Workflow
前置:确认工作目录为 wiki 项目根(包含 wiki/、raw/、tools/ 的目录)。
Step 1: 初始化
- 定位 artifact:
- 若为 slug:按顺序在
wiki/ideas/、wiki/experiments/、wiki/claims/、wiki/outputs/、wiki/papers/中查找{slug}.md - 若为文件路径:直接读取
- 记录 artifact 类型和路径
- 若为 slug:按顺序在
- 读取当前内容:加载 artifact 完整文本
- 初始化追踪变量:
round = 0score_history = []fixed_issues = []unresolved_issues = []wiki_changes = []
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.
- 7d ago First seen · 231 lines · 35 tokens per session scan A 262c95b86ab3
refine is a skill published in the GitHub repository Lambenthan/empiricalwiki (82 stars, last pushed 2mo ago), licensed MIT. It adds 35 tokens to every session and 2,614 once invoked, about $0.0002 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
paper-workflow
A staged workflow for producing an evidence-based economics or social-science research paper, from choosing a topic through submission. It coordinates existing tools and analysis options, including Python, Stata, and R.
chinese-documentation
A Chinese technical-documentation style guide covering spacing, punctuation, numbers, terminology, and links when Chinese and English appear together.
seven-pass-review
Mechanize Pattern 15 — the seven-pass adversarial review protocol for academic manuscripts. Spawns 7 forked subagents in parallel (abstract, intro, methods, results, robustness, prose, citations), then synthesizes a prioritized revision checklist. Use for submission-ready or R&R-stage papers where single-pass review…
slide-excellence
Multi-agent comprehensive slide review (visual + pedagogy + proofreading, plus TikZ / parity / substance conditionally). Use when user says "full review", "excellence pass", "comprehensive check", "review everything", "pre-release review", "slide excellence", or before teaching / shipping a deck. Fanout wrapper — for…
triage-inbox
Triage academic email and calendar (Gmail / Google Calendar via the session's MCP) into a prioritized digest plus a referee-obligations tracker — classifying referee requests, R&R and editor correspondence, co-author threads, seminar and conference invites, and grant/admin deadlines, and proposing a human-gated action…
scaffold-exercises
Scaffold a graded problem set with sections, problems, worked solutions, and short "why this matters" explainers across analytical, empirical, and coding types. Use when user says "make a problem set on X", "scaffold exercises for this lecture", "create practice problems", "generate homework with a solution key"…