CtrlK
BlogDocsLog inGet started
Tessl Logo

sn-da-image-caption

图片理解与数据提取 skill。当图片文件(.png/.jpg/.jpeg/.gif/.webp/.bmp)是主要输入且用户需要理解、提取数据或分析图片内容时使用。提供预配置的 caption 脚本(scripts/caption.py),通过 vision 模型将图片转为文本描述,无需额外配置 API Key。覆盖:(1) 通过 scripts/caption.py 对图表/表格/截图/流程图进行 caption,(2) 将 caption 文本解析为结构化 DataFrame,(3) 基于提取数据重新生成可视化图表,(4) 导出为 Excel/CSV。**遇到以下任一情况就主动使用本 skill,不要自行猜测图片内容**:①用户出现触发词:图片分析 / 图表提取 / 表格识别 / OCR / 图片描述 / 截图分析 / 图表数据 / 提取图片中的数据 / 图片转表格 / 识别图片 / image caption / extract data from image / chart analysis / table OCR;②用户上传或指定了图片文件(.png / .jpg / .jpeg / .gif / .webp / .bmp)并要求理解、提取数据或分析内容;③任务需要从图表截图、表格截图、UI 截图、流程图中提取结构化信息;④用户要求将图片中的数据转为 Excel/CSV 或重新生成可视化图表。仅不用于:图片编辑(裁剪、滤镜、缩放)、图片生成、不含数据的风景/人物照片描述。

74

Quality

92%

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

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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

A well-structured, highly actionable body that walks a complete caption→parse→visualize→export workflow with executable code and validation/recovery notes. Its only weakness is mild redundancy between the CLI and Python invocation sections, which slightly inflates token use.

Suggestions

Consolidate the 'Calling from Python' section so it only shows patterns not already covered by the CLI Usage block (e.g., parsing stdout into a DataFrame), removing near-duplicate invocations.

Trim 'What it does automatically' so it does not restate items already in the Options table, or merge the two into a single reference.

Consider moving the full parse_markdown_table and Excel-styling helpers into a bundled script or reference file to keep SKILL.md as an overview.

DimensionReasoningScore

Conciseness

Largely lean and direct, but the CLI 'Usage' examples and the 'Calling from Python' subprocess block demonstrate the same invocations, and 'What it does automatically' restates the Options table — some content could be tightened, matching the 'mostly efficient but could be tightened' anchor rather than the fully lean level 3.

2 / 3

Actionability

Provides fully executable, copy-paste-ready material throughout — concrete `caption.py` invocations with flags, a complete `parse_markdown_table` function, matplotlib font setup, and an openpyxl Excel export block — matching the 'fully executable code/commands' anchor.

3 / 3

Workflow Clarity

A clear numbered sequence (caption → parse → visualize → export) is given up front, and 'Common Pitfalls' supplies explicit validation ('Verify extracted data — check sums, percentages, row counts') and an error-recovery loop ('Large tables truncate — caption in two passes: 前半部分 + 后半部分'), matching the 'clear sequence with explicit validation and feedback loops' anchor.

3 / 3

Progressive Disclosure

Single bundle file scripts/caption.py (verified present and referenced one level deep, no nested references) plus well-organized sections; for a single-script skill this is appropriately structured with no monolithic wall of text or deep reference chains.

3 / 3

Total

11

/

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.

A highly specific, well-scoped description that concretely names capabilities, provides abundant natural trigger terms, explicitly covers both what and when, and sharply delineates its scope via exclusions. It is long but every part earns its place for trigger accuracy.

DimensionReasoningScore

Specificity

Lists multiple concrete actions — '对图表/表格/截图/流程图进行 caption', '将 caption 文本解析为结构化 DataFrame', '基于提取数据重新生成可视化图表', '导出为 Excel/CSV' — matching the 'lists multiple specific concrete actions' anchor.

3 / 3

Completeness

Clearly states what the skill does and gives an explicit 'when' clause — '遇到以下任一情况就主动使用本 skill,不要自行猜测图片内容' with four numbered trigger conditions plus explicit exclusions, matching the 'clearly answers both what AND when' anchor.

3 / 3

Trigger Term Quality

Extensive natural user-facing trigger terms in both Chinese and English (图片分析 / 图表提取 / 表格识别 / OCR / 截图分析 / image caption / extract data from image / chart analysis / table OCR), giving good coverage of what a user would actually say.

3 / 3

Distinctiveness Conflict Risk

Clear niche (image understanding/data extraction via a vision caption script) with distinct triggers and explicit exclusions ('仅不用于:图片编辑…图片生成、不含数据的风景/人物照片描述'), making overlap with other skills unlikely.

3 / 3

Total

12

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 suspicious

Warning

Total

15

/

16

Passed

Repository
OpenSenseNova/SenseNova-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.