doc-parse

将 PDF/PPT/Excel/Word 等多格式文档解析为结构化 Markdown，并输出元数据与解析置信度，作为 RAG 与四色卡片的数据底座。

Quality

62%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./antinet-agentteams/skills/doc-parse/SKILL.md

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.

The body is a compact, well-structured worker spec with concrete I/O, a real runnable script, and a clear fallback workflow. It would benefit from one inline code/CLI example and an explicit verification step in the failure-handling loop to reach the top of the scale.

Suggestions

Add a short inline example showing the script invocation with expected stdout and a snippet of doc_parse.json output so the skill is copy-paste ready.

Make validation explicit in the workflow: add a step that checks confidence/fallback_used and routes low-confidence results to human review before marking the stage complete.

Move the longer internal-context notes (RAG 复用、四色卡片) into a separate reference file or trim them so SKILL.md stays a pure overview.

Dimension	Reasoning	Score
Conciseness	The body is lean and sectioned (Input/Output/Dependencies/Failure Handling) with no padding or explanation of concepts Claude already knows; only minor internal jargon could be trimmed.	4 / 5
Actionability	It gives a concrete executable entry point (`python skills/doc-parse/scripts/run_doc_parse.py`) and lists specific parsers with their roles, but provides no inline code example or CLI argument detail beyond the run command.	4 / 5
Workflow Clarity	The three-level fallback sequence (MinerU → PyMuPDF → pdfplumber) and failure handling (degrade, BLOCKED, chunked parsing) are clearly sequenced with explicit checkpoints; it is a batch-ish operation but validation is implicit via confidence/fallback_used rather than an explicit verify step, keeping it just below 5.	4 / 5
Progressive Disclosure	Content is well organized into short sections with the runnable script and output path referenced one level deep (scripts/run_doc_parse.py, examples/.../doc_parse.json), both of which are real bundle/repository files; minor gaps are the absence of separate reference files for advanced usage.	4 / 5
	Total	16 / 20 Passed

Description

50%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.

The description is accurate and names concrete outputs (structured Markdown, metadata, confidence) but omits an explicit 'when to use' trigger clause and lacks natural trigger synonyms/extensions. It is a solid middle-of-scale description that reads more like an internal worker spec than a user-invocable skill description.

Suggestions

Add an explicit trigger clause, e.g. '当需要把 PDF/PPT/Excel/Word 等文档转成结构化文本或为 RAG 建立索引时使用'.

Include natural file extensions and synonyms (.pdf, .pptx, .xlsx, .docx, 表格、幻灯片、文档解析) to improve trigger-term coverage.

Trim internal jargon (密卷房、四色卡片) from the description and keep it focused on user-facing capabilities to reduce conflict risk with sibling skills.

Dimension	Reasoning	Score
Specificity	The description names the multi-format domain and concrete actions ('解析为结构化 Markdown，并输出元数据与解析置信度'), but stays at a high level without enumerating the distinct parsing actions per format.	3 / 5
Completeness	It clearly states what the skill does, but there is no 'Use when...' / '当...时使用' trigger clause; the when is only weakly implied via the separate assign_when field rather than the description itself, capping completeness at 3 per the rubric.	3 / 5
Trigger Term Quality	It lists format names (PDF/PPT/Excel/Word) which are natural terms, but offers no common synonyms or file extensions (e.g. .pdf, .pptx) and is framed in Chinese domain jargon (RAG, 四色卡片) rather than user-facing triggers.	3 / 5
Distinctiveness Conflict Risk	The multi-format parsing niche is somewhat specific, but '把任意格式的原始文档转成结构化文本' is broad enough to overlap with other document/extraction skills, and the internal jargon (密卷房/通政司) limits clear external distinctiveness.	3 / 5
	Total	12 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: anbeime/skill
Path: antinet-agentteams/skills/doc-parse/SKILL.md
Commit: d9e888d

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.