thesis-tracker

Buy-side discipline system. For each holding, maintain a written investment thesis — core thesis in 5 sentences, falsifiable assumptions, red lines, valuation anchors — and re-check it each quarter against new earnings/events. Scores thesis health 1-10 from assumption breakage and red-line triggers, and recommends hold / add / trim / exit. Use it right after buying a stock (to write the thesis) and every earnings season (to check it). Bridges the research→buy→monitor gap that most investors skip.

Quality

91%

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

Quality

Content

88%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 tightly structured two-mode workflow with concrete, executable guidance throughout — exact paths, tool invocations, templates, a scoring formula, and score-to-action thresholds. Its weaknesses are minor: some motivational padding (quotes, philosophy rationale) and a monolithic single-file layout where a small reference file could hold the long red-line/philosophy detail.

Suggestions

Trim the Li Lu, Keynes, and Duan Yongping quotes and compress the Design Philosophy section into one or two lines — the failure modes it lists are inferable and the quotes consume context without adding instruction.

Move the extended macro-trend red line (#6) and its explanatory rationale block into a short reference file (e.g. references/red-lines.md) to keep SKILL.md as a lean overview.

The A0 data-collection step lists tools but no fallback if financial_rigor or market data is unavailable; add one line on what to do when valuation verification fails.

Dimension	Reasoning	Score
Conciseness	The body is largely efficient — tables, templates, and a formula instead of prose — but includes padded material: the Li Lu/Keynes/Duan Yongping quotes, and a Design Philosophy section explaining investor psychology Claude can infer. Not the 5 anchor (every token does not earn its place), well above the 3 anchor (padding is minor relative to the concrete instruction density).	4 / 5
Actionability	Guidance is fully executable for an agentic workflow: exact file paths ('reports/{company}-thesis.md'), named tool calls ('web_search', 'financial_rigor (command=verify_valuation)'), fill-in templates for the 5-sentence thesis and anchor tables, an explicit health formula ('10 - (broken × 3) - (impaired × 2) - (weakening × 1) - (red-lines × 5)'), and a score-to-action table. Matches the 5 anchor for an instruction-only skill.	5 / 5
Workflow Clarity	The workflow has an explicit mode-decision checkpoint (thesis file exists → Build vs Track), clearly sequenced steps (A0-A5, B1-B7), defined statuses, and explicit red-line handling with required actions. Not a destructive/batch skill, so no validation cap applies; sequence and checkpoints match the 5 anchor.	5 / 5
Progressive Disclosure	The single SKILL.md is well-organized with clear headers and a coherent single-process scope, and no bundle files exist to mis-signpost. However, at ~160 lines the example red-line table, quotes, and philosophy rationale could partially live in a reference file; structure is good but not the ideal split of the 5 anchor.	4 / 5
	Total	18 / 20 Passed

Description

92%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: it states concrete capabilities comprehensively, gives explicit 'use when' triggers in natural language, and occupies a distinct niche (post-buy monitoring rather than research or screening). Voice is third-person imperative throughout, consistent with the good examples. Only minor gap: a few natural trigger synonyms are absent.

Dimension	Reasoning	Score
Specificity	The description lists multiple concrete actions with specifics: 'maintain a written investment thesis — core thesis in 5 sentences, falsifiable assumptions, red lines, valuation anchors', 're-check it each quarter', 'Scores thesis health 1-10', and 'recommends hold / add / trim / exit'. Coverage is comprehensive across the build and track lifecycle, matching the 5 anchor.	5 / 5
Completeness	Both parts are explicit: 'what' is fully stated (maintain thesis with named components, quarterly re-check, 1-10 health scoring, hold/add/trim/exit recommendation) and 'when' is explicit with concrete trigger phrases ('Use it right after buying a stock (to write the thesis) and every earnings season (to check it)'). Matches the 5 anchor.	5 / 5
Trigger Term Quality	Natural triggers are present ('right after buying a stock', 'every earnings season', 'each holding'), but a few common user phrasings are missing — e.g. 'should I sell', 'thesis check', 'position review'. Not the 5 anchor (no synonym-level saturation), clearly above the 3 anchor (keywords are natural and well-chosen).	4 / 5
Distinctiveness Conflict Risk	The niche — post-buy thesis discipline and quarterly re-checks of holdings — is distinct, with triggers (buying a stock, earnings season) that would not fire for research, screening, or general finance skills. Minimal conflict risk, matching the 5 anchor; it is not the 4 anchor because no meaningful overlap with closely related skills exists.	5 / 5
	Total	19 / 20 Passed

Validation

93%

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 — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: HKUDS/Vibe-Trading
Path: agent/src/skills/thesis-tracker/SKILL.md
Commit: 251b094

Reviewed: 3 days 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.