Skip to content
New

DeepSeek V4.1 Flash free for 7 days: 10M tokens a day after a top-up of $10 or more. Until Nov 30.

How it works
aeondave/libfuzzerReverse engineering and fuzzing

libfuzzer

⁨Auth/lab ref: In-process, coverage-guided fuzzing engine integrated with Clang/LLVM. For fast unit-level fuzz targets, parser hardening, sanitizer-first bug discovery, and corpus-driven regression loops in C/C++ code.⁩

v2026.11.13-21cd5683 installs+3 in 7 days39 StarsNCSAsynced 11 hours ago

Install

No NotHoly CLI yet? Install it first.

$ curl -fsSL https://notholy.com/install.sh | sh
$ notholy skills install aeondave/libfuzzer

Writes to ⁦~/.notholy/skills⁩ after checking the signature and the SHA-256.

SignatureSigned, ed25519
PGK09Vnma8…Ln0u6yAw==The CLI checks the SHA-256 and the signature before it writes a file.
Package9a57207d…7b8e6577
9a57207d01df…36d47b8e65771 file · ⁦1.4 KB⁩
ScanLow1 finding
  • Links to outside domains
See the full scan
Sourcegithub.com/AeonDave/malskill/tree/main/offensive-tools/fuzzing/libfuzzer@21cd568

LLVM-native in-process fuzzer using LLVMFuzzerTestOneInput entrypoints.

Quick Start

bash
# Minimal target
# extern "C" int LLVMFuzzerTestOneInput(const uint8_t *Data, size_t Size)

# Build with libFuzzer + ASAN
clang++ -g -O1 -fsanitize=fuzzer,address fuzz_target.cc -o fuzz_target

# Run
./fuzz_target corpus/

Operator Flow

  1. Build a narrow, deterministic LLVMFuzzerTestOneInput harness.
  2. Add ASAN/UBSAN first; add MSAN when dependency hygiene allows.
  3. Seed with small valid/invalid corpus.
  4. Run long enough to stabilize coverage features, then merge/minimize corpus.
  5. Keep crashing artifacts as regression tests.

Key Practice

  • Keep fuzz target deterministic, fast, and side-effect-light.
  • Prefer narrow targets per format/API instead of one giant dispatcher.
  • Seed with small valid+invalid samples; then minimize via -merge=1.

Useful Flags

Flag Purpose
-runs=N Bounded runs
-max_total_time=N Time budget
-dict=file Token dictionary
-jobs / -workers Parallel campaigns
-fork=N Crash/OOM/timeout-resistant subprocess mode
-use_value_profile=1 Stronger cmp-guided search
-merge=1 Corpus minimization/merge

Practical Tricks

  • Use FUZZING_BUILD_MODE_UNSAFE_FOR_PRODUCTION to disable fuzz-hostile randomness/checksums in harness builds.
  • Return -1 for intentionally rejected inputs to reduce corpus pollution.
  • Use -artifact_prefix / fixed artifact paths in CI pipelines for predictable collection.
  • For very large corpora, use resumable merge workflows.

Common Pitfalls

  • Original authors shifted major innovation to Centipede; libFuzzer remains supported for bug fixes.
  • Excellent fit for CI regression with saved crash artifacts.
  • Non-deterministic harness behavior causes corpus bloat and noisy triage.
  • Overly broad max lengths can waste cycles on low-value regions.
  • Neglecting sanitizers in in-process mode delays detection of latent memory issues.

Resources

Files in this version

Permissions

PermissionDeclaredFound by the scan
Reads filesNoNo
Writes filesNoNo
Runs shell commandsYesYes
Uses the networkYesYes
Uses connectorsNoNo
Drives a browserNoNo

Scan findings

Reviewed on Nov 13, 2026

  1. InfoLink · Links to outside domains

A clean scan lowers the risk but does not prove a package is safe. Read what you install.

VersionPublishedCommitSHA-256SignaturePermission change
v2026.11.13-21cd568Nov 13, 202621cd5689a57207dSigned–

Ask anything your scope allows.

Create an account, add credits and ask the first question in the chat or the terminal.

  1. 1/3

    Create the account

    Individual or company. You can switch to a company later without losing balance or keys.

    IndividualCompany
  2. 2/3

    Add credits

    By card. Pay per token, no subscription.

    $5.00minimum deposit
  3. 3/3

    Ask

    In the chat, or install the CLI and run notholy login in your terminal.

    $ notholy login