agentleFS
longkou1988longkou1988/cnki-skills/skills/cnki-evidence-table/SKILL.md

cnki-evidence-table

Skill28 starsChanged 2 months agoMITView on GitHub

Build an evidence-linked Excel comparison of CNKI empirical papers, including samples, variable measurement, identification, robustness and mechanisms; distinguish abstract evidence from full-text reading.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 4 are distinctive, and 2 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.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use cnki-paper-detail for source reading and cnki-download only when downloads are authorized.

    line 9
  2. Very specific:

    Read the shared ledger schema, then use ../cnki-resume/scripts/workflow.py.

    line 12
  3. Very specific:

    When the user requires full-text reading, initialize with required_access: full_text.

    line 17
  4. Distinctive:

    All included papers remain visible, even if not yet read.

    line 35
  5. Distinctive:

    Do not claim causal effects merely because the authors report correlations.

    line 38
  6. Distinctive:

    Do not upload PDFs or the private ledger to the repository or third-party services.

    line 39

≈ “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

实证研究对照表

Use for “比较这几篇论文的数据、变量、内生性处理和机制检验,附原文页码”。 Use cnki-paper-detail for source reading and cnki-download only when downloads are authorized. Reading supplied local PDFs does not require another download.

  1. Read the shared ledger schema, then use ../cnki-resume/scripts/workflow.py. Reuse the current task. For a user-selected set without earlier screening, initialize a task with selection criteria “用户指定文献”; import and mark each paper included with its observed title as evidence. Do not invent a separate exclusion exercise. When the user requires full-text reading, initialize with required_access: full_text; partial evidence stays blocked until fulfilled.
  2. Confirm title/author identity against each source. Read accessible material; record full_text, preview or abstract_only. Scanned/unreadable PDFs are a blocker until readable content is available, not completed full-text reading.
  3. Extract supported fields from the schema. Every non-empty field requires value, a short exact quote, locator, and source. Prefer PDF page number (1-based), printed page if different, table number and section. For abstracts use locator 摘要 and only facts explicitly stated there. Do not infer sample size, coefficients, causal identification or mechanisms from a title.
  4. Missing fields stay null. Distinguish “未报告” (checked the relevant full-text sections) from “未提取/证据不足” (not known). Record author's claims separately from your evaluation; field limitations contains author-stated limitations. Preserve model/column, units and sample context when recording an effect size. The program validates evidence structure, not whether a quote supports a claim; the agent must check every value against the source before saving extract.
  5. Save each paper immediately and export 研究工作表.xlsx. The 实证研究对照表 sheet holds comparisons; 原文证据 holds per-field source excerpts and locators. All included papers remain visible, even if not yet read. Report the number with completed extraction, the number read in full, and missing evidence.

Do not claim causal effects merely because the authors report correlations. Do not upload PDFs or the private ledger to the repository or third-party services.

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.