CtrlK
BlogDocsLog inGet started
Tessl Logo

nature-reader

Build full-paper Chinese-English side-by-side, figure/table-aware, source-grounded Markdown readers for journal or conference papers from PDF, DOI, arXiv, publisher HTML, or pasted text. Use whenever the user asks to translate or read a paper, make 中英文对照/原文对照/全文翻译解读, extract figures or tables into the right positions, preserve figure/table placement near relevant prose, or keep exact source anchors for every block. This skill must not degrade into a summary-only output unless the user explicitly asks for a summary. Also trigger on general paper-reading and translation requests even without the word "Nature", such as reading/translating an academic paper, literature reading, understanding a paper, and Chinese phrasings like 读论文、精读论文、论文翻译、文献翻译、文献阅读、学术阅读、帮我读这篇文章、翻译这篇paper.

62

Quality

75%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/nature-reader/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

50%

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

The router is well-organized and concise, with a clear 5-step protocol and a sensible progressive-disclosure intent. Its main weakness is that the core static-layer files and manifest.yaml it depends on are missing from the bundle, leaving its actionable guidance and verification checklist undeliverable.

Suggestions

Ship the missing core bundle the router depends on — manifest.yaml and the static/ layer (core/principles.md, core/workflow.md, core/output-contract.md) plus nature-shared/terminology-ledger.md — or inline their essential content so every referenced path resolves to a real file.

Inline the pre-response verification checklist (or at least its key checks) in SKILL.md so the workflow has an explicit validate->fix->retry feedback loop instead of deferring it to an absent file.

Trim the 'Why this split' section, which restates the static/dynamic framing already covered by the routing protocol, to improve token efficiency.

DimensionReasoningScore

Conciseness

The body is mostly lean and assumes Claude's competence (no over-explanation of PDF/DOI concepts), but the 'Why this split' section and the static/dynamic framing partly restate the routing protocol already described above, so it could be tightened; not level 3 because of that redundancy, not level 1 because there is no concept padding.

2 / 3

Actionability

Steps are concretely worded ("Read [manifest.yaml]", "Read the file mapped for the detected source_format") but the files they point to — manifest.yaml, static/core/principles.md, core/workflow.md, core/output-contract.md, ../nature-shared/core/terminology-ledger.md — do not exist in the bundle, so the promised executable guidance is missing key details; not level 3 (not copy-paste deliverable) and not level 1 (the steps are specific in form rather than vague).

2 / 3

Workflow Clarity

A clear 5-step sequence exists with one explicit checkpoint ("State the detected value in one short line... so they can correct you cheaply"), but the pre-response verification checklist is deferred to the absent core/output-contract.md rather than inlined, so the validation/feedback loop for this output-producing operation is missing per the destructive/batch-operations guideline, capping the score at 2 rather than 3.

2 / 3

Progressive Disclosure

The on-demand references that exist (references/figure-extraction.md, output-spec.md, grounding-rules.md, article-anatomy.md) are well-signaled and one level deep, but the core routing targets the body 'always_load's — manifest.yaml and the entire static/ and nature-shared/ layers — are absent from the bundle, so a large share of signaled references resolve to nonexistent files, breaking navigation; not level 3 because of these dead-end references and not level 1 because real, organized structure is present.

2 / 3

Total

8

/

12

Passed

Description

100%

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

The description is strong: third-person voice throughout, concrete multi-action capability statement, an explicit 'Use whenever...' trigger, and broad natural trigger-term coverage in both English and Chinese. It clearly carves a distinct niche with low conflict risk.

DimensionReasoningScore

Specificity

Names multiple concrete actions ("Build full-paper Chinese-English side-by-side...Markdown readers", "extract figures or tables into the right positions", "preserve figure/table placement near relevant prose", "keep exact source anchors for every block") across five distinct input sources, matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Explicitly answers both 'what' (build figure/table-aware source-grounded bilingual readers from PDF/DOI/arXiv/HTML/pasted text) and 'when' with an explicit "Use whenever the user asks to translate or read a paper..." clause, satisfying the top anchor; the 'when' is not merely implied, ruling out level 2.

3 / 3

Trigger Term Quality

Covers natural phrasings a user would actually say in both English ("translate or read a paper", "reading/translating an academic paper", "literature reading", "understanding a paper") and Chinese (读论文、精读论文、论文翻译、文献翻译、文献阅读、帮我读这篇文章), matching the 'good coverage of natural terms' anchor rather than the partial-coverage level 2.

3 / 3

Distinctiveness Conflict Risk

The full-paper bilingual-reading + source-grounding + figure/table-placement niche with specific multilingual triggers is clearly distinguishable from generic translation or document skills and unlikely to fire for the wrong skill; the slight overlap with plain 'translate' is narrowed by the full-paper/source-ground framing, keeping it at the top anchor rather than level 2.

3 / 3

Total

12

/

12

Passed

Validation

87%

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

Validation14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

relative_links

Relative link issues: 1 missing

Warning

Total

14

/

16

Passed

Repository
Yuan1z0825/nature-skills
Reviewed

Table of Contents

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.