ljg-paper

Paper reader for non-academics. Reconstructs a paper on three independent axes: the reader can identify its research problem and product, experience the evidence-driven change in understanding, and rebuild the mechanism, central experiment, and boundary. USE WHEN the user shares an arXiv link, paper URL, PDF, local paper, paper title, or asks to read, explain, analyze, or understand a paper. Defaults to a saved Org note. NOT FOR reproduction, formal peer review, benchmark tables, or literature surveys.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/ljg-paper/SKILL.md

Quality

Content

48%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The content has a sound multi-gate workflow and bundles a working validator plus templates, but it is dominated by verbose, abstract prose that restates one contract repeatedly and lacks a concrete worked example. A referenced file (ReadingGuide.md) is missing from the bundle.

Suggestions

Add one short worked example showing a filled template.org and paper-map.md for a sample paper, so the abstract three-axis contract becomes concrete and copy-pasteable.

Resolve the dangling ReadingGuide.md reference: either add the file to references/ or remove the citations and inline whatever guidance it would carry.

Collapse the repeated restatements of the three-axis contract (intro, 三重成品合同, Gotchas, Quick Reference) into a single canonical statement plus a checklist, cutting substantial token weight while preserving the gates.

Dimension	Reasoning	Score
Conciseness	The body is dense, abstract Chinese prose that spends many tokens restating the same three-axis contract in different forms (intro, 三重成品合同, Gotchas, Quick Reference) without adding executable information, assuming little of Claude's intelligence.	2 / 5
Actionability	Concrete guidance exists (the validate command, Denote filename format, 'date +%Y%m%dT%H%M%S', five-question and three-sentence gates) but most of the body is abstract instruction with no worked example of an actual Org note or filled paper-map, leaving key steps to inference.	3 / 5
Workflow Clarity	A clear sequenced workflow exists with explicit validation checkpoints — run the validator, read back via Emacs/Denote/org-lint, then pass narrative/identity/understanding gates before delivery — though the gates are described abstractly rather than as discrete ordered commands.	4 / 5
Progressive Disclosure	Structure references real bundle files (references/template.org, references/paper-map.md, scripts/validate_note.ts) and routes by input, but it repeatedly cites ReadingGuide.md which is not present in the bundle, leaving navigation with a dangling reference and no inline one-level-deep signaling of where each contract is fulfilled.	3 / 5
	Total	12 / 20 Passed

Description

100%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong, third-person description that clearly states what the skill does, when to use it via concrete natural triggers, and where it does not apply. It is specific, comprehensive, and well-bounded with minimal overlap risk.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions across three axes — 'identify its research problem and product', 'experience the evidence-driven change in understanding', 'rebuild the mechanism, central experiment, and boundary' — giving comprehensive, non-generic coverage of what the skill produces.	5 / 5
Completeness	Explicitly answers both what ('Reconstructs a paper on three independent axes...') and when ('USE WHEN the user shares an arXiv link, paper URL, PDF...') with concrete trigger phrases, plus a 'NOT FOR' exclusion clause.	5 / 5
Trigger Term Quality	Covers the natural trigger phrases users say including 'arXiv link, paper URL, PDF, local paper, paper title' and the verbs 'read, explain, analyze, or understand a paper', including file-format synonyms.	5 / 5
Distinctiveness Conflict Risk	Clear niche (paper reading for non-academics, Org note output) with distinct triggers and an explicit 'NOT FOR reproduction, formal peer review, benchmark tables, or literature surveys' boundary, minimizing conflict risk.	5 / 5
	Total	20 / 20 Passed

Validation

100%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: lijigang/ljg-skills
Path: skills/ljg-paper/SKILL.md
Commit: 44f4105

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.