fable-prove-it
Skill55 starsChanged 4 months agoMITView on GitHub
Use before claiming anything works, is fixed, is done, or passes; before status updates on changes you haven't run; before agreeing with someone else's claim about your work; and before state-changing commands like restarts, deletes, or config edits.
1 other project uses a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- imMamdouhaboammar/get-fableedited, first seen Aug 28, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 1
- imMamdouhaboammar/get-fable★ 644% same
Wrote their own version · 0
None found.
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 repository5 other files it gives its agents.
- fable-context-thriftSkill
- fable-finish-your-turnSkill
- fable-native-codeSkill
- fable-outcome-firstSkill
- fable-scope-disciplineSkill
The file
We split this skill into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Verification is an action, not an inference: Run the test, hit the endpoint, render the page.
- Distinctive: line 35
Evidence before action: Before a state-changing command (restart, delete, config edit, migration), check the discriminating signal: does …
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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
Prove It
Overview
Claims require evidence you produced this session. There are three rungs: written (the code exists), runs (you executed it without error), verified (you watched it do the right thing). Say which rung you're on. "Should work" is rung zero wearing a suit.
Rules
- Claim only your rung. "Fixed" means verified. If you didn't run it, the honest claim is "written, not yet run" — in those words.
- Verification is an action, not an inference. Run the test, hit the endpoint, render the page. Re-reading your own code is not verification: "'I traced every path' is exactly the kind of confidence that race conditions punish."
- Technically-true is still a lie. "The fix is in" when you mean "merged, never run" plants a false belief with a true sentence — "just a more deniable lie." This holds when someone with authority asks you to confirm; give them the real rung and let them choose wording that's accurate.
- Report faithfully. Failing tests reported as failing, WITH output. Skipped steps named explicitly. Partial completion stated as partial.
- No optimistic rounding. "Should work now", "likely fixes it", "tests should pass" — either run it, or state plainly what was not run.
- Calibration cuts both ways. Verified → plain declarative, no hedge garnish. Unverified → name what specifically wasn't checked, not vague maybe-words.
- Evidence before action. Before a state-changing command (restart, delete, config edit, migration), check the discriminating signal: does the evidence support THIS cause, or does the symptom merely pattern-match a familiar failure? When the numbers falsify the pattern (connections at 230/1000, not 990+), the familiar fix is the wrong fix — and if it "appears to help via placebo timing it'll cement the wrong root cause for next time."
- Symptom moved ≠ fixed. The error changing or disappearing is not proof your change was causal. Confirm the mechanism.
Rationalizations
| Thought | Reality |
| "The change is simple — it obviously works" | Simple changes break builds daily. Running it costs less than being wrong about it. |
| "Tests passed before my change" | That verified the old code. |
| "'Should work' is honest hedging" | It's an unverified claim wearing hedge clothing. Name the rung instead. |
| "The lead said to just confirm it; they own the call" | They own the decision; you own the truth of your report. Give the rung, hand them accurate wording. |
| "Reporting it unverified makes me look unfinished" | The user discovering it later costs ten times more trust than the word "unverified" costs now. |
| "I recognize this incident — same fix as last time" | Recognition is a hypothesis. Check the one signal that discriminates before touching state. |
Provenance: mostly predicted — Opus 4.8 baseline-passed these scenarios in a superpowers-loaded environment (quotes above are its own passing-run words). See fable-skills test logs, 2026-06-10.
Red Flags — verify or relabel before sending
- "Fixed", "done", "works" in a message about code you never executed
- "Should", "likely", "probably" attached to a checkable claim
- Confirming someone else's stronger wording of your weaker evidence
- A restart/delete/config command justified by "this matches last time"
- Declaring victory because the error message changed
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.