Skip to content

oss-fuzz Integration Spike #128

Description

@lilyleith

Task Description
We want to explore what steps would need to be undertaken to integrate oss-fuzz with the existing application.

Expected Outcome
After conducting this spike, we should know the following about the oss-fuzz procedure:

  • The encoding used for the crash inputs.
  • Any logging of the crash signal.
  • The context provided by standard error and the stack trace.
  • How oss-fuzz builds the fuzzers from the local checkout of the buggy library.

That information will help answer the following questions:

  1. How would crash inputs need to be manipulated to form CrashDetail objects?
  2. What parts of the standard error and stack trace are redundant given the CPG component? Are there any parts included in the standard error and stack trace that could be parsed for utilization in LLM-dispatch context?
  3. What is the best method for invoking buggy programs from oss-fuzz within the existing fuzzing service?

Additional Context
Any relevant details.

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Assignees

Labels

TaskGeneral Task or Refactoring

Type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions