6verity

数学建模竞赛最终验证和验收阶段，支持 Typst 和 LaTeX 双引擎。用于论文写完后检查章节数量、标题顺序、图表引用、数值一致性、占位符、内部文件泄露、参考文献、代码可复现性、编译和提交就绪状态。

Quality

61%

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/6verity/SKILL.md

Quality

Content

63%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 is a well-sequenced, actionable 9-step verification workflow with real executable commands and explicit validation/pass criteria appropriate to a destructive acceptance gate. Its weaknesses are moderate verbosity, inlined check-list detail that could be externalized, and weakly-signaled file references.

Suggestions

Move the long Typst/LaTeX figure- and section-matching check lists into a reference file or rely on the script output, keeping SKILL.md as an overview, and signal references with markdown link syntax.

Tighten repeated guidance (path-hardcoding warnings, xelatex two-pass rationale) to a single mention to improve token efficiency.

Add explicit 'only proceed when the previous step passes' checkpoints between every step to push workflow clarity toward 5.

Dimension	Reasoning	Score
Conciseness	The body is mostly efficient and task-focused with executable bash blocks, but it restates some obvious guidance (e.g. explaining that xelatex needs two passes, restating 'do not hardcode paths' in multiple places) and the section/figure-matching check lists are somewhat verbose; could be tightened.	3 / 5
Actionability	Provides concrete executable bash for the writing check script, typst/xelatex compilation, and PDF-to-PNG export with real commands and flags; a few placeholders (e.g. SCRIPT_PATH and $PAPER_DIR) are explicitly justified as needing runtime resolution, which is acceptable but leaves minor gaps.	4 / 5
Workflow Clarity	A clear 9-step sequence with explicit validation gates (script FAIL = hard error must fix and rerun, compilation must succeed before PDF check, PASS criteria enumerated) and feedback loops for error recovery; this is a destructive/batch verification skill and validation is present, so it clears the cap. Minor gaps (e.g. no explicit 'only proceed to Step N when Step N-1 passes' for every step) keep it below 5.	4 / 5
Progressive Disclosure	Structure is reasonable with staged sections and a real referenced file (scripts/writing_check.sh) plus one one-level-deep norm reference (../_references/math_modeling_norms.md), but the body inlines a large amount of check-list content that arguably belongs in the script or a reference doc, and inline reference signaling is weak (no markdown link syntax, paths given as plain backticks).	3 / 5
	Total	14 / 20 Passed

Description

58%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 specific and well-differentiated, naming many concrete verification actions and a clear domain (math modeling competition paper acceptance, Typst/LaTeX). Its main weakness is the absence of an explicit 'Use when...' trigger clause and limited natural-language trigger terms, which caps completeness and trigger quality.

Suggestions

Add an explicit trigger clause such as 'Use when a math modeling competition paper is finished and needs final verification/acceptance before submission.'

Surface natural trigger terms users would say, e.g. '验证论文', '检查论文', '论文验收', '编译论文', alongside the existing domain vocabulary.

Optionally mention the PDF visual check in the capability list to round out coverage toward score 5.

Dimension	Reasoning	Score
Specificity	Lists several concrete actions (check chapter count, title order, figure references, numerical consistency, placeholders, internal file leakage, references, reproducibility, compilation, submission readiness) with broad but real coverage; minor gaps (e.g. no mention of PDF visual check) keep it below 5.	4 / 5
Completeness	Clearly states 'what' (final verification/acceptance stage that checks a list of items) but 'when' is only weakly implied by '论文写完后' (after the paper is written); there is no explicit 'Use when...' / trigger clause, which per guidelines caps completeness at 3.	3 / 5
Trigger Term Quality	Contains relevant domain terms (数学建模竞赛, Typst, LaTeX, 论文) and concrete check categories, but lacks common natural trigger phrases users would actually say (e.g. '验证论文', '检查论文', '验收', '编译失败') and synonyms; mostly domain-jargon rather than user-spoken keywords.	3 / 5
Distinctiveness Conflict Risk	The combination of 数学建模竞赛 final acceptance with Typst/LaTeX dual-engine and the specific check list carves a clear niche distinct from general writing or compilation skills; only minor overlap risk with a generic LaTeX-compile skill.	4 / 5
	Total	14 / 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
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning

	Total	15 / 16 Passed

Repository: jihe520/MathModelAgent
Path: skills/6verity/SKILL.md
Commit: 83d8783

Reviewed: 12 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.