cnki-researcher
Coordinate a CNKI literature workflow from search and screening to evidence-based reading and citation export.
longkou1988 · 已有静态检查记录
安装与使用
运行命令前先阅读来源与依赖说明。复制命令不会自动安装或执行。
npx skills add longkou1988/cnki-skills --skill cnki-researcher技能说明
name: cnki-researcher description: Coordinate a CNKI literature workflow from search and screening to evidence-based reading and citation export.
CNKI research workflow
Use the sibling cnki-* skills when installed; read each before its stage. If one is missing, report the missing module and perform only stages supported by the available tools. These are instructions, not automatic subagent spawning.
- State the database, query/fields, period, sort, requested count and citation scope. Default to Chinese journal articles. For “since 2023” use 2023-01-01 through today.
- Use cnki-search or cnki-advanced-search, cnki-navigate-pages and cnki-parse-results. Freeze a unique, source-linked result list before screening; retain site rank.
- Select the requested subset based on title/abstract relevance to the question. Explain selection and retain original list numbers. Do not substitute newer or better-known papers outside the list without clearly reporting the change.
- Use cnki-paper-detail. Full-text analysis requires retrieved readable full text; otherwise label abstract/preview analysis and identify missing evidence. When downloads are requested, use cnki-download with PDF as the default. Create the local Downloads/CNKI/YYYY-MM-DD folder first (user's local date; respect a supplied destination), then save and verify files there. Reuse the current session and follow cnki-download's bounded progress-based waiting. Preserve an explicit user format choice; otherwise do not substitute CAJ when PDF is unavailable. Continue accessible items and report the shortfall.
- Use cnki-export for the requested citation set. If “export the citations” is unspecified, export the listed set and a separate selected-subset file.
- Before delivery compare all titles/identifiers against extracted records, ensure no duplicate filler entries, and verify exported counts and file paths.
- Return the search manifest, result list, reading analysis, .bib artifact and any limitations. Never claim a blocked stage completed.
Durable screening, comparison and resume
When the user requests screening tables, empirical comparisons, or resumable work, use cnki-resume from task start and save each verified result immediately. Use cnki-screening for criterion-linked include/exclude/uncertain decisions, and cnki-evidence-table for per-field evidence with page/section locators. Follow their shared schema; use one task file through these stages. Existing local readable papers can be analyzed without downloading again. Missing sibling modules are reported, never implicitly installed.
On continuation, read the existing queue before any web action. Reuse completed records and verified files; inspect running transfers before retry. Preserve the original query/criteria and selection scope. Export the workbook even if access blocks some records, while explicitly reporting unfinished work and abstract-only analysis. Do not claim structural evidence validation proves scholarly accuracy.
Browser and access
Optional paid screening: use cnki-jev only when installed and explicitly enabled for this task with an authorized title/abstract upload scope and budget. Read its SKILL.md before use. Otherwise keep the baseline workflow. Jev proposals do not replace frozen criteria, source evidence, or the existing screening ledger.
Use the user's available browser connector. Claude Code: Chrome DevTools MCP
(list pages → select page → take snapshot → interact). Codex: use its available
browser skill/tool and documented API. WorkBuddy: the browser-skill CLI (bsk)
— bsk session start → bsk navigate <url> --session <id> → bsk snapshot →
interact → bsk session stop <id> (mandatory, even on error paths). Read that
tool's instructions first; do not assume Chrome DevTools names exist in the other
runtimes. If no connector is available, report the missing dependency, without
installing it implicitly.
Start at an already-open CNKI official/authorized institutional entry. Otherwise default to the domestic CNKI site, in this order:
- https://kns.cnki.net/kns8s/defaultresult/index — domestic 检索 entry (verified working: renders 主题 + 学术期刊 controls and the result table, and honours an institutional login).
- https://www.cnki.net/ — domestic homepage, only if the entry above is unreachable.
Never start at an overseas mirror (oversea.cnki.net, global.cnki.net,
data.oversea.cnki.net, scholar.cnki.net, ar.oversea.cnki.net or similar).
Detection: if the page title is "CNKI Overseas" or the host contains oversea,
you are on the international platform — www.cnki.net may silently redirect there
depending on the network exit point. Navigate straight to the kns.cnki.net entry
above instead of staying on the mirror.
Preserve the current institution route and observed links; do not reconstruct
proxy URLs or trust a hostname solely because it contains "cnki".
Inspect current visible controls before acting; refresh observations after navigation.
Do not invent selectors, search API payloads, article URLs, download hashes or session tokens.
Use read-only DOM extraction only if supported and grounded in observed markup.
If an operation fails twice without progress, preserve completed records and report
the blocker. Login and CAPTCHA require user intervention under the runtime policy;
do not evade detection or bypass access controls. Do not export cookies or credentials.
Close owned temporary tabs when finished, retaining only a page needed for user handoff.
Overseas fallback (last resort only)
Fall back to an overseas mirror only when the domestic site is genuinely unreachable or blocked for the user — for example kns.cnki.net times out, returns a network/ISP block page, or shows an access blocker with no institutional route available. When that happens:
- Tell the user explicitly that the domestic site failed, state the observed reason (timeout, block page, missing institution e