agentleFS
Xuzhen-LiXuzhen-Li/literature-review-skills/skills/journal-selection/SKILL.md

journal-selection

Skill2 starsChanged 35 days agoMITView on GitHub

Builds a shortlist of 3-5 journals that publish reviews, checked on each journal's official site and dated. Use when the user says "pick a journal", "shortlist journals", "where should we submit", "impact factor", "APC", or "lock the journal". Do not use to invent a journal title, copy a metric from a third-party ranking site, or submit the manuscript.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

9 rules. About 0 are common, 6 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Do not put a quartile or an impact-factor number in the table or the fit cell.

    line 55
  2. Very specific:

    Write <project>/11-journal/journal_shortlist.md with the table from references/shortlist_template.md.

    line 57
  3. Distinctive:

    No tracking parameters. The date is the day you opened the page, not a date in the future.

    line 61
  4. Very specific:

    After the user locks the choice, record it with the gate name in shared/gates.md (fallback name journal_lock) without deleting other keys, …

    line 85
  5. Very specific:

    Every recommended journal was opened on an official author page, and the verified sentence matches that URL and today's date.

    line 100
  6. Distinctive:

    No impact factor, APC, quartile, or turnaround number was copied from a third-party ranking, a matcher site, or a search snippet.

    line 101
  7. Distinctive:

    Do not invent ISSN, APC, impact factor, acceptance rate, or "time to first decision".

    line 108
  8. Distinctive:

    Do not describe a journal as MEDLINE-indexed, DOAJ-listed, or "Q1" unless the official list was opened and the date is in the file.

    line 109
  9. Distinctive:

    Do not submit, create an account, or write to an editor in this skill.

    line 110

≈ “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

Journal selection

Opening

On first use in a project, say this verbatim:

I am journal selection. I shortlist journals from their official pages and I copy a metric only from a page opened today.

When to use

A full draft exists, or the user has asked for a journal before drafting and accepts that manuscript-polishing will still have to fit the locked journal's limits. Dynamic facts (scope, article types, APC, word limit, indexing, turnaround, impact factor) are written only after you open an official page, and each fact is dated.

Inputs (paths) / Preconditions

  • <project>/project.yaml — field, audience, review type.
  • <project>/07-outline/outline.md — target word count and article type.
  • <project>/06-exemplars/style_brief.md when present.
  • <project>/evidence/evidence_table.csv — the venue column named in shared/evidence_table_schema.md, for included reviews.
  • The user's own journal names, if they gave any.

<suite> and <project> are the suite directory and the review folder.

Procedure (numbered steps)

  1. Build the candidate list only from three places: names the user gave, journals that published the exemplars in 06-exemplars/, and journals that published the review articles already in the evidence table. If that list has fewer than three real titles, ask the user for more names and stop. Do not add a title from memory, from a ranking blog, or from a search snippet.
  2. For each candidate, open the publisher's author-guidelines page for that journal. Record the URL. Ignore Wikipedia, ResearchGate, library LibGuides, LetPub, SciRev, and social networks. Indexing and metrics have their own official pages, listed in references/official_sources.md.
  3. Copy these facts from the page you opened, or write the token shown when the page does not state them:
    • article type name as printed (Review, Systematic review, or whatever the page calls it)
    • accepts_reviews: yes or no
    • invited_only: yes or no
    • OA model and APC, with the currency as printed, or [VERIFY]
    • word limit, or [VERIFY]
    • reference style, or not stated
    • turnaround, or not stated (do not import a number from a third-party site)
    • whether review is double-anonymized
  4. If the user wants a citation metric, including an impact factor or a JCR quartile, or a national partition, record it only from an official page opened that day. Use the journal's own page or Clarivate Journal Citation Reports (https://jcr.clarivate.com/) for a citation metric, including a JCR quartile. For a national partition, open the official list the user named and copy that page's own label. Do not open LetPub, SCImago, journal finders, or other matcher sites. Add the lines under the table, or omit them when the user did not ask:
Citation metric for Journal Name: 4.2 (2024 Journal Impact Factor)
Metric source for Journal Name: verified on YYYY-MM-DD at https://jcr.clarivate.com/...
National partition for Journal Name: [label copied from the official list]
Partition source for Journal Name: verified on YYYY-MM-DD at https://...

Do not put a quartile or an impact-factor number in the table or the fit cell. A JCR quartile is a citation metric, not a national partition. If you did not open the page, omit the lines. Do not write them from memory. 5. Indexing: if the author page claims MEDLINE, Scopus, Web of Science, or DOAJ, confirm on the official list in references/official_sources.md and date that check too. If you did not open the list, the indexing cell is [VERIFY]. 6. Write <project>/11-journal/journal_shortlist.md with the table from references/shortlist_template.md. Three to five data rows. The verified cell is exactly:

verified on YYYY-MM-DD at https://the-author-guidelines-url

No tracking parameters. The date is the day you opened the page, not a date in the future.

  1. Under the table, add:
Recommended: Journal Name
Backup: Other Journal Name
Double-anonymized: no
Reference style: Vancouver

Recommended and Backup must be different rows, both with accepts_reviews = yes. If invited_only is yes, add Invitation for Journal Name: yes only when the user has an invitation. Otherwise do not recommend that journal.

  1. Run:
python3 <suite>/skills/journal-selection/scripts/check_shortlist.py \
  <project>/11-journal/journal_shortlist.md

Fix every ERROR until it prints OK shortlist.

  1. Show the user the recommended journal, the backup, the word limit, the APC or [VERIFY], and the guidelines URL. Do not edit project.yaml yet.
  2. After the user locks the choice, record it with the gate name in shared/gates.md (fallback name journal_lock) without deleting other keys, and append:

YYYY-MM-DD | journal-selection | journal lock passed: Journal Name | human

Human gate(s)

Journal lock. A passing checker is not the lock. If the user refuses, append journal lock refused | human and stop. manuscript-polishing does not start without the lock. An optional decision card may come first (shared/decision_cards.md): four labeled options, one preferred, and a custom option rewritten as a plan with its risks. Only the exact sentence I confirm the journal lock sets the lock.

Outputs (exact paths + formats)

  • <project>/11-journal/journal_shortlist.md — the 12-column table plus the four lines in step 7. Optional partition and citation-metric lines are included only when the user asked and the source page was opened that day.

Quality checklist

  • The checker prints OK shortlist.
  • Every recommended journal was opened on an official author page, and the verified sentence matches that URL and today's date.
  • No impact factor, APC, quartile, or turnaround number was copied from a third-party ranking, a matcher site, or a search snippet.
  • A partition or citation metric has a same-day official source line, or it is absent.
  • Invited-only journals are not recommended unless the invitation line is present.
  • The file contains 3 to 5 journals, not a longer shopping list.

Guardrails

  • Do not invent ISSN, APC, impact factor, acceptance rate, or "time to first decision".
  • Do not describe a journal as MEDLINE-indexed, DOAJ-listed, or "Q1" unless the official list was opened and the date is in the file.
  • Do not submit, create an account, or write to an editor in this skill.

Next step

exemplar-review-analysis to refresh the style brief from lawful reviews in the locked journal, then manuscript-polishing. If the locked word limit is below the outline target, polishing cuts to the limit and logs every cut. It does not reopen the science.

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.