CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

patent-reader

中国专利通俗解读:公开号/PDF 成通俗笔记、图谱与 Obsidian 入库。

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Passed

No findings from the security scan

Fix and improve this skill with Tessl

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

Quality

Content

75%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.

A lean, densely operational orchestrator skill with a clear step sequence, conditional branches, and named validation checkpoints. Its critical weakness is progressive disclosure: all delegated detail files are missing from the actual bundle while the bundle's real reference and asset files are never surfaced, making the skill hard to navigate and its delegated steps unexecutable as bundled.

Suggestions

Fix the reference tree: prompts/patent_plain_reader.md, prompts/type_hooks.md, prompts/fill_*, obsidian_ofm_companion.md, patent_reader_self_check.md and the whole tools/ directory do not exist in the bundle — either restore them or repoint the steps to files that do exist.

Surface the existing bundle files in the body — e.g. references/patent_pdf_sources.yaml, references/schemas/claim_features.schema.yaml (directly relevant to producing claim_features.json), and assets/patent_note_template.md — so they are discoverable from SKILL.md.

Add invocation examples with arguments for the three scripts and an explicit validate → fix → re-check loop around validate_claim_features.py to raise workflow clarity to anchor 5.

DimensionReasoningScore

Conciseness

The ~30-line body is pure operational content — directives, exact file paths, output specs — with zero padding and no explanation of concepts Claude already knows. Every token earns its place, matching the 'lean and efficient' anchor.

5 / 5

Actionability

Steps name exact files to Read (prompts/patent_plain_reader.md, prompts/type_hooks.md) and per-purpose scripts (tools/extract/fetch_patent_pdf.py, tools/vault/write_patent_obsidian_note.py, tools/analyze/validate_claim_features.py), which is mostly executable guidance. It falls short of anchor 5 because no invocation examples with arguments are given, so the commands are not copy-paste ready.

4 / 5

Workflow Clarity

The numbered 步骤 section has a clear sequence with conditional branches (对照表派工 stops early; 实用新型/外观 adds type hooks) and names validation tools (validate_claim_features.py, patent_reader_self_check.md). Not 5 because there is no explicit validate → fix → re-check feedback loop; the checkpoints are named but not wired into the sequence.

4 / 5

Progressive Disclosure

Scored against the actual bundle: every path the body delegates to — prompts/patent_plain_reader.md, prompts/type_hooks.md, prompts/fill_*, obsidian_ofm_companion.md, patent_reader_self_check.md, and the entire tools/ tree — is absent from the bundle, so navigation does not resolve. Meanwhile the files that do exist (references/patent_pdf_sources.yaml, references/schemas/claim_features.schema.yaml, assets/patent_note_template.md) are never mentioned in the body, leaving them undiscoverable. Not 3 because the references are not merely unclearly signaled but broken in both directions.

2 / 5

Total

15

/

20

Passed

Description

70%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 compact, concrete description with a clearly distinct niche: it names the domain, the input forms (publication number, PDF), and three concrete outputs (plain notes, graphs, Obsidian ingestion). Its main weakness is the complete absence of a 'use when' trigger clause, which caps completeness, plus a few missing natural trigger synonyms.

Suggestions

Add an explicit trigger clause, e.g. 「用户说读专利/读懂这个专利、给出公开号或专利 PDF 时使用」, so completeness is no longer capped at 3 by the missing 'use when' guidance.

Include natural trigger variations users would actually say — 专利号, 专利文献, .pdf — to broaden keyword coverage toward anchor 5.

DimensionReasoningScore

Specificity

"公开号/PDF 成通俗笔记、图谱与 Obsidian 入库" names concrete inputs (publication number, PDF) and three specific outputs (plain-language notes, graphs, Obsidian ingestion), matching the 'several specific actions with minor gaps' anchor. Not 5 because key capabilities like claim-feature extraction and machine-readable outputs are unmentioned; not 3 because it goes beyond 1-2 actions.

4 / 5

Completeness

"中国专利通俗解读:公开号/PDF 成通俗笔记、图谱与 Obsidian 入库" gives a clear, concrete 'what', but there is no 'use when' clause or equivalent explicit trigger guidance, capping this dimension at 3 per the judging guidelines. Not 4 because the 'when' is entirely absent rather than merely underspecified.

4 / 5

Trigger Term Quality

Natural terms a user would say are present — 中国专利, 公开号, PDF, 通俗解读 — giving good keyword coverage. A few common variations are missing (专利号, 专利文献, .pdf), so it sits between anchor 3 and the comprehensive synonym coverage of anchor 5.

4 / 5

Distinctiveness Conflict Risk

The niche is unmistakable — Chinese patent plain-language interpretation with Obsidian vault ingestion — so it would rarely trigger for the wrong skill. Overlap risk with generic PDF or note-taking skills is minimal, matching the 'clear niche with distinct triggers' anchor.

5 / 5

Total

17

/

20

Passed

Validation

100%

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

handsomestWei/patent-disclosure-skill

Path

skills/patent-reader/SKILL.md

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.