CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

investment-memo-craft

Codex-only writing and layout overlay for AI Berkshire investment research reports. Use whenever Codex creates, rewrites, revises, or critiques company/industry/fund research reports, especially long-form Markdown reports that need financial rigor, readable business mechanics, contrarian analysis, valuation-to-action guidance, investor-specific recommendations, restrained typography, and clear buy/hold/sell signals. Do not use this to modify Claude Code slash-command sources.

Quality

80%

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

Fix and improve this skill with Tessl

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

Quality

Content

67%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 instruction-only skill with concrete title/filename templates, a fixed section ordering, and explicit audit pairing. Its weaknesses are redundancy between the Core Workflow and Default Report Shape sections and the absence of any worked example of the key output artifacts.

Suggestions

Merge the 'Core Workflow' and 'Default Report Shape' sections into one structure: each numbered step can state both what to analyze and where it lands in the report, eliminating the current step-by-step duplication.

Add one short worked example (a filled-in metadata block or a summary/action table skeleton) so the highest-leverage artifacts are shown, not just described.

Fold the overlapping emphasis rules from 'Style Standards' and 'Layout Standards' into a single bold/emphasis guideline to cut repeated guidance.

DimensionReasoningScore

Conciseness

Mostly efficient imperative prose with no conceptual padding, but it could be tightened: the 9 'Core Workflow' steps substantially duplicate the 11 'Default Report Shape' sections (moat -> 护城河评估, inverse thinking -> 逆向思考, management -> 管理层评估, valuation -> 估值), and 'Style Standards' vs 'Layout Standards' repeat emphasis guidance ('Use bold sparingly...' vs 'Keep ordinary facts in normal weight'). Not 2 because there is no explanation of concepts Claude already knows and every bullet states a standard; not 4 because the workflow/shape overlap is a genuine ~30-40 line redundancy, not a minor trim.

3 / 5

Actionability

Concrete, copy-paste-ready formats: exact title pattern '公司名(ticker)研究报告', dated filename '公司名研究报告-YYYYMMDD.md', fixed Chinese step headings, an 11-section ordered template, and a Quality Bar question checklist. Not 5 because there is no worked example of the metadata block, summary table, or action table — the highest-leverage artifacts are described but never shown; not 3 because the guidance is specific and executable, not pseudocode-level hints.

4 / 5

Workflow Clarity

Clear 9-step sequence with an explicit ordering rule ('Put the detailed recommendation ... near the end, after business quality, risk, and valuation have been argued'), a checklist ('Quality Bar'), and validation checkpoints delegated to named tools ('cross-source validation', 'report_audit.py before treating a report as publishable'). Not 5 because validation lives in sibling skills rather than an integrated validate-fix-retry loop, and what to do when report_audit.py fails is unspecified.

4 / 5

Progressive Disclosure

Single file with no bundle directories, so there is no nesting risk; the body is well-sectioned (Purpose, Core Workflow, Style, Layout, Default Report Shape, Quality Bar, Pairing) making navigation easy, and references to sibling skills are explicit. Not 5 because at ~150 lines the Layout Standards and Default Report Shape detail is a candidate for a one-level reference file, and no 'see X for more' structure exists at all; not 3 because the in-file organization is good and nothing is buried.

4 / 5

Total

15

/

20

Passed

Description

83%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.

A strong description: third-person, explicit 'Use whenever...' trigger with concrete verbs, a clear niche, and a negative boundary clause. The main gaps are the slightly abstract 'overlay' framing of what it does and missing common synonyms like 'memo' and 'thesis'.

DimensionReasoningScore

Specificity

Lists several concrete actions ('creates, rewrites, revises, or critiques' research reports) plus specific output attributes ('contrarian analysis, valuation-to-action guidance, ... clear buy/hold/sell signals'), but 'writing and layout overlay' is somewhat abstract about the actual transformations performed. Not 5 because it never states the concrete mechanics (moat scoring, price bands, report sectioning) that the body delivers; not 3 because the action verbs and capability list go well beyond 1-2 generic actions.

4 / 5

Completeness

Explicitly answers both: what ('writing and layout overlay for AI Berkshire investment research reports' with a enumerated capability list) and when ('Use whenever Codex creates, rewrites, revises, or critiques company/industry/fund research reports, especially long-form Markdown reports'), with a concrete negative boundary ('Do not use this to modify Claude Code slash-command sources'). Matches the anchor-5 example structure of what + 'Use when...' trigger phrases.

5 / 5

Trigger Term Quality

Good natural keyword coverage: 'research reports', 'company/industry/fund', 'investment', 'valuation', 'buy/hold/sell signals', 'rewrites', 'critiques'. Not 5 because common synonyms a user might say ('memo', 'thesis', 'due diligence', 'analysis') are absent, and 'AI Berkshire'/'Codex-only' are jargon-ish; not 3 because coverage clearly exceeds a single generic keyword.

4 / 5

Distinctiveness Conflict Risk

Clear niche ('AI Berkshire investment research reports', 'Codex-only') with an explicit conflict exclusion for Claude Code slash-command sources. Not 5 because 'creates, rewrites, revises, or critiques company/industry/fund research reports' overlaps the sibling research skills (financial-data, investment-research) it is designed to pair with, and the description alone doesn't fully disambiguate the overlay-vs-underlying-research boundary.

4 / 5

Total

17

/

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.

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository

xbtlin/ai-berkshire

Path

codex-skills/investment-memo-craft/SKILL.md

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.