Journal Latest Issue

Source: https://github.com/aipoch/medical-research-skills

Validation Shortcut

Run this minimal command first to verify the supported execution path:

python scripts/journal_digest.py --help

When to Use

• You want the latest issue table of contents (TOC) for one or more journals for quick scanning.
• You need a structured digest (title/DOI/abstract) plus Chinese key points for internal reading notes without using external translation services.
• You want to export results to Markdown and CSV for review, sharing, or archiving.
• You have an RSS feed or TOC URL and want a repeatable pipeline to pull the newest issue content.
• You need a batch mode (JSON input) to process multiple journals in one run.

Key Features

• Fetches the latest issue TOC and (optionally) abstracts by crawling public web pages only.
• Supports multiple input sources: journal name, RSS, TOC URL, and batch JSON configuration.
• Generates Chinese key points locally based on title + abstract (no external translation APIs); keeps the original English abstract for verification.
• Outputs:
  • output.md (human-readable digest grouped by journal/issue)
  • output.csv (structured table for downstream processing)
• Safety controls:
  • Domain allowlist by default; optional override via --allow-all
  • Built-in timeout/retry behavior (implementation-dependent)

Dependencies

• Python 3.x
• Python standard library only (no third-party packages required)

Example Usage

1) Single journal (optionally provide RSS/TOC)

python scripts/journal_digest.py --journal "Nature" --rss "https://example.com/rss"

2) Batch mode via JSON

Create input.json:

{
  "journals": [
    { "name": "Nature", "rss": "https://...", "toc": "https://..." }
  ]
}

Run:

python scripts/journal_digest.py --json input.json

3) Common optional flags

python scripts/journal_digest.py \
  --journal "Nature" \
  --max-items 20 \
  --fetch-abstracts

Implementation Details

Inputs

Supported CLI inputs:

• --journal: Journal name (repeatable)
• --rss: RSS/TOC feed URL (optional)
• --toc: Table of Contents page URL (optional)
• --json: Batch input file (JSON)

Batch JSON format:

{
  "journals": [
    { "name": "Journal Name", "rss": "https://...", "toc": "https://..." }
  ]
}

Optional parameters:

• --allow-all: Allow crawling non-whitelisted domains (must be explicitly enabled)
• --max-items: Maximum number of items to output (default: 20; set to 0 for unlimited)
• --fetch-abstracts: Additionally crawl abstracts from article landing pages (slower)

Source selection logic

• If --rss or --toc is provided, the skill prioritizes those sources to identify the latest issue content.
• If neither is provided, the skill may attempt built-in source hints for common journals (e.g., known RSS endpoints). For best accuracy, prefer providing RSS/TOC (and/or stable identifiers such as ISSN if supported by your setup).

Chinese key points generation (no external translation APIs)

• The skill does not call external translation services.
• It produces Chinese key points as a brief outline derived from the English title and abstract.
• The original English abstract is retained in outputs for manual review and validation.

Outputs

• output.md: Latest issue directory and generated key points, grouped by journal/issue.
• output.csv: Tabular export (e.g., Title / DOI / Abstract / Chinese Key Points).

Security and compliance notes

• Network access is limited to crawling public pages.
• A domain allowlist is enforced by default; use --allow-all only when you explicitly accept the risk of crawling unknown domains.
• Timeout/retry mechanisms are used to improve robustness (exact values depend on the implementation).

When Not to Use

• Do not use this skill when the required source data, identifiers, files, or credentials are missing.
• Do not use this skill when the user asks for fabricated results, unsupported claims, or out-of-scope conclusions.
• Do not use this skill when a simpler direct answer is more appropriate than the documented workflow.

Required Inputs

• A clearly specified task goal aligned with the documented scope.
• All required files, identifiers, parameters, or environment variables before execution.
• Any domain constraints, formatting requirements, and expected output destination if applicable.