retest
Skill4 starsChanged 7 days agoMITView on GitHub
Re-score a business idea after the user has run their test. Loads a previous idea-jury report, checks the pre-set kill criteria against the new results, and either says stop or puts the idea back before a (quick or full) jury with the new evidence. Updates the idea ledger so the verdict history is visible. Use when the user says "retest", "update my idea", "here are my results", "I ran the test", "I talked to N customers", "it didn't work", "re-score", or refers to an earlier jury report or idea slug.
2 other projects use a version of this.
Adopted by
Most recently taken up by
- ratomirjovanovic/pandawardtheir own version, first seen Nov 1, 2026
- mindfortai/security-skillstheir own version, first seen Jun 29, 2026
Other versions2 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 2
- ratomirjovanovic/pandaward★ 50same name
- mindfortai/security-skills★ 16same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repositoryOne other file it gives its agents.
- jurySkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 32
If the user's message has no concrete numbers, ask once for: how many real buyers they spoke to, how many said yes to the price, how many …
- Distinctive: line 42
If any stop line was hit, say so plainly and recommend the action the founder wrote down (stop or pivot).
- Very specific: line 50
If no stop line was hit, update the Case File with the new facts: tag results [S] (the founder's own data), update interviews: and …
- Distinctive: line 68
Stop lines are decided in advance and honoured.
- Distinctive: line 70
If the evidence is thin (under 5 conversations, no payments), say the retest cannot move the verdict much and what would.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
retest
Evidence changes verdicts. This skill makes sure the founder follows the stop lines they wrote down before the test, and re-scores only after that.
Input: $ARGUMENTS
Step 1: Load
Find idea-jury-reports/<slug>-*.md (newest if several) and
idea-jury-reports/ledger.csv. If the slug is missing or ambiguous, list the
ledger rows and ask which idea, in one line.
Extract from the old report: the verdict, the kill criteria sheet, the what would have to be true table, and the budget.
Step 2: Get the results
If the user's message has no concrete numbers, ask once for: how many real buyers they spoke to, how many said yes to the price, how many paid or pre-paid and how much, hours and money actually spent, and anything that surprised them. Do not accept "it went well" as a result.
Step 3: Check the kill criteria first
For each kill criterion, compare the threshold with the actual result and mark it HIT, MISSED or NOT YET TESTABLE. Show this as a table.
- If any stop line was hit, say so plainly and recommend the action the founder wrote down (stop or pivot). Do not reinterpret the numbers or invent reasons to continue. The founder may still decide otherwise, but the report must record that they overrode their own criterion.
- Compare actual hours and money with the budget range.
Step 4: Re-score
If no stop line was hit, update the Case File with the new facts: tag
results [S] (the founder's own data), update interviews: and
anyone_has_paid:, and cross out assumptions that are now known. Do not
include the previous verdict, so the jurors stay independent. Then run the
jury skill's process on the updated case file (use the quick jury unless the
user asks for the full one), including the scorer.
Step 5: Report the delta
Show: old verdict, new verdict, what changed it (which assumption moved from
[A] or [U] to [S]), what is still unknown, and the next smallest test
with new kill criteria and a new review date.
Append a row to idea-jury-reports/ledger.csv for the retest and save the
new report as idea-jury-reports/<slug>-<date>-retest.md.
Rules
- Stop lines are decided in advance and honoured. That is the point.
- A GO still requires buyer evidence, which the founder's test may now supply.
- If the evidence is thin (under 5 conversations, no payments), say the retest cannot move the verdict much and what would.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

