thesis-tracker

Compares existing investment theses against new data to classify evidence and flag rating changes.

Updated May 9, 2026
One-click install
npx skills add https://github.com/122yjs/invest_harness --skill thesis-tracker-122yjs
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: thesis-tracker
Source: https://github.com/122yjs/invest_harness/tree/main/plugins/vertical-plugins/invest-research/skills/thesis-tracker
Command: npx skills add https://github.com/122yjs/invest_harness --skill thesis-tracker-122yjs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Investment theses go stale as new earnings, events, and market data arrive, and manually checking whether each original claim still holds is slow and error-prone. This Skill structures that review so every claim is re-evaluated against fresh evidence with clear sourcing. ## Core Features & Use Cases - Claim Decomposition: Breaks an existing investment thesis into 3-5 discrete claims and records the original evidence behind each one. - Evidence Classification: Labels new data as strengthening, maintaining, weakening, or unverified for each claim, and marks unsourced claims as unverified. - Rating Impact & Tracking Plan: Notes whether a Rating or Price Target recalculation is needed and defines the next tracking indicators and review cadence. - Use Case: After a company reports quarterly earnings, feed the prior report and the new results into this Skill to produce a thesis-update.md file showing which claims still hold and what to monitor next. ## Quick Start Ask the assistant to run the thesis tracker for a given ticker using the existing report and the latest earnings event, writing the result to the workspace thesis-update file.

Frequently Asked Questions about thesis-tracker

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check if an investment thesis still holds after new earnings?▼

Provide the ticker, the existing thesis or report path, and the new earnings data. The Skill decomposes the thesis into 3-5 claims, classifies the new data as strengthening, maintaining, weakening, or unverified for each claim, and writes the result to thesis-update.md.

What inputs does an investment thesis tracking workflow need?▼

It needs a company name or ticker, the existing thesis or prior report path, the new data or event with a time-period hint, and an active workspace path. Missing or unsourced claims are marked as unverified rather than assumed.

Can thesis tracking automatically change a stock rating or price target?▼

No. It only flags whether a Rating or Price Target change is possible and notes which outputs must be recalculated. Final change decisions are left to QA and valuation re-validation steps.

What happens when a claim has no supporting source?▼

Claims without a source are explicitly labeled as unverified evidence. The Skill does not merge old thesis content with new data to produce definitive conclusions without proper grounding.

Where is the thesis update output written?▼

The output is written to thesis-update.md under the 05_macro_sentiment directory of the active workspace, keeping it consistent with the broader file-handoff research pipeline.