CtrlK
BlogDocsLog inGet started
Tessl Logo

kimi-datasource

Universal data-source assistant for stocks (Wind, S&P, SEC EDGAR), macro (World Bank, IMF, FRED, NBS), Chinese government data and standards (GB/HB/DB/TT), corporate, academic, legal, WHO/FAO/OECD and other IGO data, financial news (Xinhua, Caixin). This plugin exposes tools via MCP server `plugin-kimi-datasource_data`; call them in the flow `mcp__plugin-kimi-datasource_data__get_data_source_desc` → `mcp__plugin-kimi-datasource_data__call_data_source_tool`.

70

Quality

85%

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

SKILL.md
Quality
Evals
Security

Quality

Content

96%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 exceptionally lean and actionable, with a well-sequenced workflow that includes explicit validation and error-handling checkpoints for risky operations like stock-code verification and file_path handling. The only mild gap is progressive disclosure: everything is inlined into one ~167-line file with no bundle references, which is appropriate for a single-purpose MCP skill but leaves the long capability table unsplit.

DimensionReasoningScore

Conciseness

The body is lean and assumes Claude's competence — it deliberately refuses to copy API names/params ("这份 skill 故意不抄具体的 API 名和参数表"), delivers every token as actionable guidance, and avoids explaining what a data source or CSV is, matching the score-5 lean-and-efficient anchor.

5 / 5

Actionability

It gives fully executable, copy-paste-ready guidance: exact MCP tool names, concrete JSON params (e.g. {"name":"stock_finance_data"}, {"ticker":"600519.SH","start_date":"...","file_path":"/tmp/mao_1y.csv"}), file_path conventions, batch-size limits, and worked examples, matching the score-5 fully-executable anchor.

5 / 5

Workflow Clarity

The standard workflow is a clearly numbered sequence with explicit validation checkpoints: read desc before calling, verify stock codes via web_search ("调任何股票相关 API 前,先用 web_search 核对"), use full company names, stop when results cover the question, and report errors rather than retrying — matching the score-5 anchor with feedback loops for risky operations.

5 / 5

Progressive Disclosure

Structure is good with clear section headers (调用方式, 能力, 工作流, 铁律, 读返回结果, watchlist, 注意事项) and the large 25-source capability table is appropriately inlined as the skill's core map; however no bundle files exist and the long reference-style capability/boundary tables could plausibly live in a separate file, so it sits at score-4 rather than 5.

4 / 5

Total

19

/

20

Passed

Description

73%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 trigger-term rich, clearly naming the data domains and the exact MCP tool-call flow, but it omits an explicit "Use when..." clause that tells Claude when to invoke it. Adding a one-sentence usage trigger would lift completeness.

Suggestions

Add an explicit "Use when..." trigger clause naming the natural situations a user would prompt (e.g. "Use when the user asks for stock quotes, macro indicators, Chinese standards, academic papers, or financial news that requires a registered data source").

Trim the opening word "Universal" which slightly broadens the niche and increases overlap risk; the enumerated source list already establishes the scope.

Voice is third person ("exposes", "call them"), which is correct; keep it as you refine.

DimensionReasoningScore

Specificity

The description names the broad domain ("stocks (Wind, S&P, SEC EDGAR), macro (World Bank, IMF, FRED, NBS), Chinese government data and standards (GB/HB/DB/TT)...") plus concrete actions ("exposes tools via MCP server", "call them in the flow get_data_source_desc → call_data_source_tool"), giving several specific capabilities with minor coverage gaps, matching the score-4 anchor.

4 / 5

Completeness

It has a clear "what" (the listed data sources and the MCP tool flow) but lacks an explicit "Use when..." / "when should Claude use it" trigger clause; per the rubric a missing Use-when clause caps completeness at 3.

3 / 5

Trigger Term Quality

It lists comprehensive natural terms users would actually say — "Wind", "S&P", "SEC EDGAR", "World Bank", "IMF", "FRED", "NBS", "GB/HB/DB/TT", "Xinhua", "Caixin" — plus concrete tool names, matching the comprehensive-coverage anchor for score 5.

5 / 5

Distinctiveness Conflict Risk

The enumerated source list and specific MCP tool names give it a clear, mostly-distinct niche with only minor overlap risk against a generic financial-data skill, fitting the score-4 anchor; not a 5 because the opening word "Universal data-source assistant" is somewhat broad.

4 / 5

Total

16

/

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.

Validation16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository
MoonshotAI/kimi-code
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.