# US 2026/0093601 A1 · Generating static defect checkers using a language model

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2026/0093601 A1
- Document ID: US20260093601A1
- Status: Published application
- Kind: A1
- Application number: 18/900,628
- Filed: 2024-09-27
- Published: 2026-04-02
- Earliest claimed: 2024-09-27
- Assignees: Oracle International Corporation
- Applicants: Oracle International Corporation
- Inventors: Cristina CIFUENTES; Isaac BEH; Kirra FLETCHER; Jonathan FUNG
- Web page: https://patentgrep.com/patent/US20260093601A1

## CPC

- G06F 11/3604: Analysis of software for verifying properties of programs (testing of software G06F11/3668)
- G06F 11/3608: {using formal methods, e.g. model checking, abstract interpretation (theorem proving G06N5/013)}
- G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N)

## Abstract

A method generates static defect checkers using a language model. The method includes generating an example representation. The method further includes combining an explanation section, an instruction section, and a description section to generate a prompt. The explanation section includes the example representation, the instruction section includes instruction text, and the description section includes defect description text. The defect description text includes a natural language description of a defect corresponding to the example representation. The explanation section includes operations corresponding to the defect. The instruction text includes instructions in the natural language to generate defect checker code using the example representation and the defect description text. The method further includes executing a language model using the prompt to generate the defect checker code. The defect checker code is in a programming language.

## Claims

### Claim 1

A method comprising:
generating an example representation;
combining an explanation section, an instruction section, and a description section to generate a prompt, wherein the explanation section comprises the example representation, the instruction section comprises instruction text, and the description section comprises defect description text, wherein the defect description text comprises a natural language description of a defect corresponding to the example representation, wherein the explanation section comprises operations corresponding to the defect, and wherein the instruction text comprises instructions in the natural language to generate defect checker code using the example representation and the defect description text; and
executing a language model using the prompt to generate the defect checker code, wherein the defect checker code is in a programming language.

### Claim 2 (depends on 1)

The method of claim 1, wherein generating the example representation comprises:
executing an intermediator using example code to generate the example representation from the example code.

### Claim 3 (depends on 1)

The method of claim 1, wherein generating the example representation comprises:
converting example code to a graph comprising a set of nodes representing operations corresponding to the example code and comprising a set of edges between the set of nodes representing execution paths within the example code; and
filtering the set of nodes to generate a filtered set of nodes corresponding to the operations corresponding to the defect.

### Claim 4 (depends on 1)

The method of claim 1, wherein combining the explanation section, the instruction section, and the description section to generate the prompt comprises:
appending the instruction section to the explanation section, wherein the explanation section comprises explanation text in the natural language describing the example representation.

### Claim 5 (depends on 1)

The method of claim 1, wherein combining the explanation section, the instruction section, and the description section to generate the prompt comprises:
appending the description section to the instruction section.

### Claim 6 (depends on 1)

The method of claim 1, wherein the defect is one or more of a lock defect, a memory allocation defect, and a tainted defect.

### Claim 7 (depends on 1)

The method of claim 1, further comprising:
executing an intermediator using test code to generate a test representation from the test code.

### Claim 8 (depends on 1)

The method of claim 1, further comprising:
executing the defect checker code using a test representation generated from test code to generate a defect report, wherein the defect report indicates a presence of the defect within the test code.

### Claim 9 (depends on 1)

The method of claim 1, further comprising:
testing the defect checker code with code samples comprising example code to generate an accuracy score of the defect checker code, wherein the code samples comprise a set of positive test cases that evaluate to true when the defect is present and a set of negative test cases that evaluate to false when the defect is present; and
preventing deployment of defect checker code after comparing the accuracy score to an accuracy threshold.

### Claim 10 (depends on 1)

The method of claim 1, further comprising:
testing a plurality of defect checker codes, comprising the defect checker code, with code samples comprising example code to generate a plurality of accuracy scores, wherein an accuracy score of the plurality of accuracy scores represents an accuracy of the defect checker code; and
selecting the defect checker code using the accuracy score.

### Claim 11

A system comprising
at least one processor; and
an application that, when executing on the at least one processor, performs operations comprising:
generating an example representation;
combining an explanation section, an instruction section, and a description section to generate a prompt, wherein the explanation section comprises the example representation, the instruction section comprises instruction text, and the description section comprises defect description text, wherein the defect description text comprises a natural language description of a defect corresponding to the example representation, wherein the explanation section comprises operations corresponding to the defect, and wherein the instruction text comprises instructions in the natural language to generate defect checker code using the example representation and the defect description text; and
executing a language model using the prompt to generate the defect checker code, wherein the defect checker code is in a programming language.

### Claim 12 (depends on 11)

The system of claim 11, wherein generating the example representation comprises:
executing an intermediator using example code to generate the example representation from the example code.

### Claim 13 (depends on 11)

The system of claim 11, wherein generating the example representation comprises:
converting example code to a graph comprising a set of nodes representing operations corresponding to the example code and comprising a set of edges between the set of nodes representing execution paths within the example code; and
filtering the set of nodes to generate a filtered set of nodes corresponding to the operations corresponding to the defect.

### Claim 14 (depends on 11)

The system of claim 11, wherein combining the explanation section, the instruction section, and the description section to generate the prompt comprises:
appending the instruction section to the explanation section, wherein the explanation section comprises explanation text in the natural language describing the example representation.

### Claim 15 (depends on 11)

The system of claim 11, wherein combining the explanation section, the instruction section, and the description section to generate the prompt comprises:
appending the description section to the instruction section.

### Claim 16 (depends on 11)

The system of claim 11, wherein the defect is one or more of a lock defect, a memory allocation defect, and a tainted defect.

### Claim 17 (depends on 11)

The system of claim 11, wherein the application performs operations further comprising:
executing an intermediator using test code to generate a test representation from the test code.

### Claim 18 (depends on 11)

The system of claim 11, wherein the application performs operations further comprising:
executing the defect checker code using a test representation generated from test code to generate a defect report, wherein the defect report indicates a presence of the defect within the test code.

### Claim 19 (depends on 11)

The system of claim 11, wherein the application performs operations further comprising:
testing the defect checker code with code samples comprising example code to generate an accuracy score of the defect checker code, wherein the code samples comprise a set of positive test cases that evaluate to true when the defect is present and a set of negative test cases that evaluate to false when the defect is present; and
preventing deployment of defect checker code after comparing the accuracy score to an accuracy threshold.

### Claim 20

A non-transitory computer readable medium comprising instructions executable by at least one processor to perform operations comprising:
generating an example representation;
combining an explanation section, an instruction section, and a description section to generate a prompt, wherein the explanation section comprises the example representation, the instruction section comprises instruction text, and the description section comprises defect description text, wherein the defect description text comprises a natural language description of a defect corresponding to the example representation, wherein the explanation section comprises operations corresponding to the defect, and wherein the instruction text comprises instructions in the natural language to generate defect checker code using the example representation and the defect description text; and
executing a language model using the prompt to generate the defect checker code, wherein the defect checker code is in a programming language.
