CtrlK
BlogDocsLog inGet started
Tessl Logo

earnings-team

AI Berkshire skill: 财报精读团队:四大师并行解读 + 公众号发布. Source: skills/earnings-team.md.

53

Quality

58%

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/earnings-team/SKILL.md
SKILL.md
Quality
Evals
Security

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.

A well-structured, highly actionable multi-agent earnings-analysis workflow with concrete commands, templates, and a real validation/exit feedback loop. Its main weaknesses are verbosity (meta adapter notes and philosophy prose) and a monolithic single-file structure with no progressive disclosure despite the length.

Suggestions

Move the four per-agent deep-dive specs (段永平/巴菲特/芒格/李录) and/or the report templates into reference files under references/ and link to them one level deep, keeping SKILL.md as an overview.

Trim or relocate the 'Codex adapter note' and '设计理念' sections to reduce token overhead unrelated to the core workflow.

Make the Team Lead synthesis checkpoints mechanical (e.g. an explicit checklist for consensus/contradiction resolution) to lift workflow clarity from mostly-checkpoints to full validation rigor.

DimensionReasoningScore

Conciseness

The ~460-line body is mostly efficient domain-specific guidance (per-agent analysis points, signal-light systems, templates) but includes padding that could be trimmed: the 'Codex adapter note', the '设计理念' philosophy prose, and the sibling-skill relationship table. Matches anchor 3 (mostly efficient, some unnecessary explanation); not 4 because several sections are genuine overhead rather than minor trimmable instances.

3 / 5

Actionability

Provides concrete, mostly-executable guidance: specific bash commands (financial_rigor.py cross-validate / verify-valuation / three-scenario, report_audit.py extract/verdict), exact report templates, signal-light rubrics, and per-agent output formats. Matches anchor 4 (concrete commands with minor gaps); the placeholders ({值1}, {公司名}) need filling and some guidance (e.g. 管理层语气分析) is qualitative, so not 5.

4 / 5

Workflow Clarity

Clear three-phase sequence (研究/合成/发布) with numbered steps, progress tracking, and an explicit validation/exit gate ('数据抽检/准出流程') with a 打回→修正→重审 feedback loop plus cross-validation during research. Matches anchor 4 (clear sequence, most checkpoints present); not 5 because the inter-agent synthesis checkpoints (consensus/contradiction resolution by Team Lead) are qualitative rather than mechanical validation steps.

4 / 5

Progressive Disclosure

The skill ships as a single ~460-line SKILL.md with no references/scripts/assets bundle; everything is inline under good section headings, but content that could live in separate reference files (the four per-agent deep-dives, report templates) is all inlined with no one-level-deep file references. Matches anchor 3 (some structure, content that should be separate is inline, no clearly signaled references); the simple-skill exception does not apply since it far exceeds 50 lines.

3 / 5

Total

14

/

20

Passed

Description

53%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 states a clear 'what' with a couple of concrete actions and natural Chinese trigger terms, but it omits any 'when to use' guidance and under-enumerates the workflow's stages. It is distinct from sibling skills but slightly generic in phrasing.

Suggestions

Add an explicit 'Use when ...' clause naming trigger situations (e.g. 重要公司关键财报需要深度解读并发布公众号文章时使用).

Enumerate more concrete actions/stages — parallel research by four analysts, synthesis, editorial rewrite, reader review — rather than only '并行解读 + 公众号发布'.

Include common trigger synonyms (财报 / earnings / 季度报 / 年报) to improve keyword coverage and reduce overlap with '/earnings-review'.

DimensionReasoningScore

Specificity

Names the domain ('财报精读团队') plus two concrete actions ('四大师并行解读' and '公众号发布'), but coverage is not comprehensive — it omits the synthesis/reader-review stages present in the body. Fits anchor 3 (domain + 1-2 concrete actions); not 2 because the actions are concrete rather than purely generic, and not 4 because multiple specific actions are not enumerated.

3 / 5

Completeness

Has a clear 'what' (team-based earnings deep-read + publishing) but no 'when' / 'Use when...' clause at all, so per rubric guidance completeness is capped at 3. Not 2 because the 'what' is clear rather than vague; not 4 because 'when' is entirely absent rather than weakly present.

3 / 5

Trigger Term Quality

Contains natural Chinese keywords users would say ('财报精读', '公众号') but misses common variations and synonyms (e.g. 'earnings', '季度报', '年报', '财报分析'). Matches anchor 3 (some relevant keywords, missing common variations); not 4 because keyword coverage is thin rather than 'good with a few missing'.

3 / 5

Distinctiveness Conflict Risk

The team + publishing framing carves a distinct niche, though the body reveals a closely related sibling skill ('/earnings-review', single-agent) creating minor overlap risk. Matches anchor 4 (mostly distinct, minor overlap with a closely related skill); not 5 because that sibling overlap exists, not 3 because the multi-agent + publishing scope is clearly differentiated.

4 / 5

Total

13

/

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.