CtrlK
BlogDocsLog inGet started
Tessl Logo

investment-research

AI Berkshire skill: 投资研究:巴菲特-芒格-段永平-李录 四大师综合分析框架. Source: skills/investment-research.md.

47

Quality

49%

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

Fix and improve this skill with Tessl

tessl review fix ./codex-skills/investment-research/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

70%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 strong, highly actionable research workflow with best-in-class sequencing and explicit validation/feedback loops throughout. Its weaknesses are length that could be trimmed and a monolithic structure that inlines material (terminal-value methodology, data-source spec) which would benefit from being split into referenced bundle files.

DimensionReasoningScore

Conciseness

The ~330-line body is dense with genuine domain methodology (r-g currency matching, C1/C2/C3 audit constraints, terminal-value model) that Claude does not trivially know, so most tokens earn their place; however the overall length and several explanatory asides ('C3为什么是硬的', the circular-reasoning quote) could be tightened without losing clarity.

3 / 5

Actionability

It provides concrete, near-copy-paste bash commands with explicit flags (verify-market-cap, cross-validate, three-scenario, terminal_value audit/pe/irr) plus filled-in argument templates, tables and checklists; the remaining gap is the {placeholder} substitution and a few instructional rather than executable passages, keeping it just below fully copy-paste ready.

4 / 5

Workflow Clarity

The eight-step process is explicitly sequenced with hard validation checkpoints and feedback loops: mandatory data cross-validation, the three-constraint audit gate ('不通过不许把估值写进报告'), and the post-write data-sampling exit flow with 打回/准出 retry-until-pass logic and checklists — a textbook match for the top anchor.

5 / 5

Progressive Disclosure

Section headers are clear and external references are signaled (skills/financial-data.md, tools/*.py, reports/*.md), but no bundle files exist and the whole methodology is inlined in one ~330-line SKILL.md; content that could live one level deep (the terminal-value methodology, the data-source spec) is not split out, so it sits at 'some structure, could be better organized'.

3 / 5

Total

15

/

20

Passed

Description

28%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 essentially a localized title plus source provenance rather than a capability statement: it names the domain but gives no concrete actions, no use-when trigger guidance, and is padded with meta-noise. It is distinguishable only by the named-masters framing, which is itself under-explained.

Suggestions

Rewrite the description in third person as concrete capabilities, e.g. 'Conducts systematic investment research on a target company using the Buffett-Munger-Duan-Masters framework: collects and cross-validates financial data, evaluates moats, models valuation with terminal-value audit, and produces a buy/hold/avoid memo.'

Add an explicit 'when' clause such as 'Use when the user asks to analyze a stock, value a company, or produce an investment research report on a named target.'

Remove the meta-noise ('AI Berkshire skill:', 'Source: skills/investment-research.md.') from the description — provenance does not help Claude decide when to invoke the skill.

DimensionReasoningScore

Specificity

The description names the domain ('投资研究综合分析框架' framed around four named masters) but lists no concrete actions beyond the generic '分析' — it reads as a title, not a capability statement, and is diluted by meta-noise like 'AI Berkshire skill:' and 'Source: skills/investment-research.md.'.

2 / 5

Completeness

It offers a vague 'what' (a four-master investment analysis framework) and no 'when' guidance at all — there is no 'Use when...' clause, which per the rubric caps completeness at 3 and the missing trigger guidance pulls it further down to the 'vague what, no when' anchor.

2 / 5

Trigger Term Quality

Only one or two generic keywords ('投资研究', '分析框架') appear; the natural phrases a user would actually say when needing this skill ('分析这家公司', '估值', '护城河') are missing, and the meta-noise crowds out trigger language.

2 / 5

Distinctiveness Conflict Risk

The four-master framing ('巴菲特-芒格-段永平-李录') is a recognizable niche, but the description reduces to a generic 'investment research analysis framework' that could overlap with other finance/analysis skills, so it sits at 'somewhat specific but could still overlap'.

3 / 5

Total

9

/

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
xbtlin/ai-berkshire
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.