Generate concrete counterexamples when formal verification, assertions, or specifications fail. Use this skill when debugging failed proofs, understanding why verification fails, creating minimal reproducing examples, analyzing assertion violations, investigating invariant breaks, or diagnosing specification mismatches. Produces concrete input values, execution traces, and state information that demonstrate the failure.
81
77%
Does it follow best practices?
Impact
85%
1.10xAverage score across 3 eval scenarios
Passed
No known issues
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/counterexample-generator/SKILL.mdStructured counterexample report
Failed Property named
100%
100%
Location identified
100%
100%
Concrete inputs given
100%
100%
Expected vs Actual
100%
100%
Step-by-step trace
41%
100%
Root Cause section
100%
100%
Root cause is correct
100%
100%
Suggested Fix section
100%
100%
Fix is correct
100%
100%
Minimal Example section
37%
100%
Regression test cases
50%
100%
5-step workflow and boundary analysis
Step-by-step process documented
100%
100%
Both bugs found
100%
100%
Violation types labeled
100%
100%
Boundary values used
100%
100%
Concrete inputs for enqueue bug
100%
100%
Concrete inputs for peek bug
100%
100%
Execution trace for at least one bug
100%
100%
Root causes explained
100%
100%
Suggested fixes included
100%
100%
Generalization present
0%
50%
Preconditions identified
100%
100%
Minimal example and generalization
truncate bug found
100%
100%
count_words bug found
100%
100%
pad_center not falsely flagged
37%
25%
Minimal example for truncate
80%
100%
Minimal example for count_words
80%
100%
Minimality verified
25%
50%
Root causes explained
100%
100%
Regression tests provided
0%
0%
Tests cover fix verification
0%
0%
Generalization section
20%
0%
Execution trace present
100%
100%
0f00a4f
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.