thesis-drift

AI Berkshire skill: 投资论文漂移检测：分清事实变化与措辞变化. Source: skills/thesis-drift.md.

Quality

67%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./codex-skills/thesis-drift/SKILL.md

Quality

Content

77%

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 thorough, actionable workflow with strong sequencing and validation, but it is a long monolithic file with some non-essential framing that could be trimmed and no progressive file structure.

Suggestions

Trim the design-rationale section and the Keynes epigraph to the minimal rule set; the three change categories can be stated in one line each.

Move the dimension-judgment tables and the report output template into a reference file (e.g. references/drift-template.md) linked one level deep from the overview.

Collapse the repeated Improved/Unchanged/Weakened column text in the drift table since the verdict vocabulary is already fixed earlier in the body.

Dimension	Reasoning	Score
Conciseness	The body is mostly actionable tables and commands, but the design-rationale section and Keynes epigraph explain context Claude could infer, and the ~225-line length could be tightened.	2 / 3
Actionability	It provides executable commands (python3 tools/financial_rigor.py verify-valuation ... with full flags), concrete judgment tables, and a copy-paste-ready report template.	3 / 3
Workflow Clarity	The workflow is clearly sequenced (mode selection → A1–A6 / B / C) with explicit validation checkpoints (numerical cross-validation, anti-mismatch pairing) and error-recovery handling for missing baselines.	3 / 3
Progressive Disclosure	The skill is a single well-organized file with clear sections but no bundle files or one-level-deep references, and detailed tables/templates that could be split out remain inline.	2 / 3
	Total	10 / 12 Passed

Description

57%

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 conveys a clear, distinct purpose in a specialized niche but reads as a label plus source citation rather than a trigger-oriented description. It omits explicit use-when guidance and broader natural trigger terms.

Suggestions

Add a 'Use when...' clause naming natural triggers, e.g. 'Use when comparing two thesis reports or snapshots and deciding whether the investment thesis has drifted.'

List concrete actions beyond the single 'distinguish fact vs wording changes', such as compare report snapshots, run valuation verification, and produce a drift report.

Drop the 'Source: skills/thesis-drift.md' provenance note from the description and fold common trigger variations (thesis drift, 报告对比) into the trigger clause.

Dimension	Reasoning	Score
Specificity	Names a concrete domain (投资论文漂移检测) and one action (分清事实变化与措辞变化), but does not enumerate multiple concrete actions like compare reports or cross-validate valuations.	2 / 3
Completeness	It states what the skill does but lacks any 'Use when...' trigger guidance, so the 'when' is missing rather than explicit.	2 / 3
Trigger Term Quality	The phrase 投资论文漂移检测 is a relevant natural term, but common variations (thesis drift, 报告对比, 论文变化) are absent and the term is somewhat niche.	2 / 3
Distinctiveness Conflict Risk	The thesis-drift niche is clearly scoped and unlikely to trigger for unrelated skills, though it depends on a separate /thesis-tracker baseline.	3 / 3
	Total	9 / 12 Passed

Validation

100%

Warnings & errors only

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/thesis-drift/SKILL.md
Commit: 634a399

Reviewed: 1 day ago

Table of Contents

Discovery Implementation Validation

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.