Put new evidence beside your original reasons and see what still holds.
See what you will get
Three original assumptions, checked against evidence.
Synthetic teaching sample · not a real company / fund
Three original assumptions, checked against evidence.
Original assumption · Aug 1, 2026
New evidence · Oct 30, 2026
Review
Double-digit revenue growth
Comparable revenue +12%
Supported this period
Operating cash flow / profit ≥80%
Current ratio 69.70%
Challenged
No material customer losses
Retention data missing
Unknown
Synthetic DEMO-CONSUMER record. The 80% condition belongs to this example user, not a platform standard. One supported, one challenged and one unknown assumption do not constitute an automatic sell instruction. The original record stays intact.
Keep the records in your workspace
Support, counterevidence, unknowns and next checkpoints
thesis-review.mdchanges.mdnext-check.md
Each run writes to a new reports/{run-id}/, preserving inputs and previous reports. Status distinguishes completed, needs-input and failed.
What to prepare
Your original dated thesis, an as-of date and new sources. No position size, cost basis or brokerage access is required.
Use the thesis check Skill to compare my original thesis with new evidence. Do not invent past reasons or edit the original file. Mark support, counterevidence and unknowns, and list the next questions to check.
First ask your host to inspect the method version, inputs, output location and dependencies in read-only mode. Downloading, discovering a Skill and completing its sample are separate steps.
Its limits
Without an old thesis, establish a new baseline rather than claiming historical drift. Save revisions separately after confirmation. Follow-up conditions remain notes, not scheduled monitoring or trades.
No network retrieval, account connection or result upload by default. Model handling depends on your host and service settings; this is not a promise of fully offline processing.
