Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →athola/justify
Audits changes for additive bias and Iron Law compliance.
Passed 15 of 21 test cases with the skill, and 6 of 21 without it.
and 12 more in the full report
10 fixed6 passed either way6 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
During a diff audit of `src/server.ts`, the git statistics show 35 added lines and 5 deleted lines (a 7:1 ratio). What signal score and interpretation applies to this ratio?
| Checked | Without skill | With skill |
|---|---|---|
| A line ratio between 5:1 and 10:1 receives a signal score of 2. | Missed | Passed |
| A signal score of 2 indicates additive bias is likely. | Missed | Passed |
The instructions your model reads when the skill loads.
The simplest change that fixes the problem is the safest change to merge. Adding code is easy. Removing the need for code is engineering.
AI models are trained to be helpful, which creates a systematic bias toward adding code rather than fixing root causes:
| AI Default Behavior | Correct Behavior |
|---|---|
| Add a workaround | Fix the root cause |
| Modify test expectations | Fix the implementation |
| Create a new helper | Use an existing one |
| Add error handling | Prevent the error |
| Add a compatibility shim | Remove the old code |
| Wrap in try/catch | Fix the exception source |
This skill audits changes for these patterns and requires explicit justification for each.
imbue:karpathy-principles)imbue:scope-guard)test cases passed, vs 6 / 21 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.