CtrlK
BlogDocsLog inGet started
Tessl Logo

thesis-tracker

AI Berkshire skill: 投资论文追踪:买入后的纪律系统. Source: skills/thesis-tracker.md.

48

Quality

51%

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/thesis-tracker/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

75%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, actionable workflow with concrete templates, commands, file paths, and a clear two-mode sequence plus validation checkpoints. Its main weaknesses are non-essential motivational/rationale padding that hurts conciseness and a description that under-sells the skill.

Suggestions

Trim the motivational quotes and the '设计理念' narrative to a one-line rationale; keep the templates and process which carry the value.

Consider moving the assumption/red-line/valuation table templates into a references/ template file linked from the body, which would improve both conciseness and progressive_disclosure for this 200+ line skill.

Add an explicit validate→fix→retry note around the verify-valuation step in A0 (e.g. 'if validation fails, correct the inputs and re-run before proceeding') to push workflow_clarity toward 5.

DimensionReasoningScore

Conciseness

The templates, tables, and process steps earn their tokens, but the motivational quotes (李录/凯恩斯/段永平), the '设计理念' rationale, and the Codex adapter note are explanation Claude does not need, fitting 'mostly efficient but includes some unnecessary explanation'; it is not 4 because the padding is more than minor.

3 / 5

Actionability

Guidance is fully executable: concrete file paths (reports/{公司名}-thesis.md), a real command (tools/financial_rigor.py verify-valuation), a 5-question thesis template, assumption/red-line/valuation tables, and an explicit health-score formula cover the common cases copy-paste ready.

5 / 5

Workflow Clarity

The two-mode flow (A0–A5 build, B1–B7 track) is clearly sequenced with checkpoints (A0 valuation validation, B3 per-assumption checks, B4 red-line checks) and a red-line→action rule; it stops at 4 because there is no hard validate→fix→retry gate of the kind the anchor-5 example shows.

4 / 5

Progressive Disclosure

No bundle files exist, and the content is cohesive templates organized into clear sections (设计理念, 执行流程, 模式A/B, 关键原则) with no need for external references, giving good single-file structure; it is not 5 because the file is well over 50 lines and some template material could plausibly live in a reference.

4 / 5

Total

16

/

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 a sparse topic label: it names the domain but gives no concrete actions, no natural trigger phrases, and no 'Use when' guidance, while risking overlap with sibling investment skills. It is usable but well below the clarity bar set by the reference examples.

Suggestions

Rewrite the description as third-person actions, e.g. 'Tracks an investment thesis after purchase: establishes the thesis, runs quarterly checks against core assumptions and red lines, and scores thesis health.', replacing the 'AI Berkshire skill' label and source citation.

Add an explicit trigger clause such as 'Use when the user wants to set up or review an investment thesis for a held position (e.g. 公司名, 公司名 建立论文, 公司名 季度检查).'

Drop the 'Source: skills/thesis-tracker.md' provenance line from the description — it is metadata, not activation signal, and adds no trigger value.

DimensionReasoningScore

Specificity

The description names the domain ("投资论文追踪:买入后的纪律系统") but states no concrete actions — it is a topic label ("AI Berkshire skill: ... Source: ...") rather than a list of what the skill does, matching the 'names the domain but actions are minimal or generic' anchor; it does not reach 3 because no actions at all are enumerated.

2 / 5

Completeness

The 'what' is a vague topic label with no described actions and the 'when' is entirely missing (no 'Use when...' clause), matching anchor 2 ('vague what and no when'); the missing trigger guidance also caps completeness at 3, and the vague what pulls it to 2.

2 / 5

Trigger Term Quality

Only one real keyword ("投资论文追踪") appears; the natural user phrases the body relies on ("建立论文", "季度检查") are absent, fitting 'one or two generic keywords; missing the natural phrases users say' rather than 3 which requires broader relevant coverage.

2 / 5

Distinctiveness Conflict Risk

The post-purchase thesis-tracking niche is somewhat specific, but the body references sibling skills /investment-research and /investment-team, indicating real overlap with closely related investment skills, fitting 'somewhat specific but could still overlap with similar skills'.

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.