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/radamsaReverse engineering and fuzzing

radamsa

⁨Auth/lab ref: General-purpose mutation engine for generating malformed test inputs.⁩

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

Install

No NotHoly CLI yet? Install it first.

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

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

SignatureSigned, ed25519
6NAR18ENcw…gzfTVrCg==The CLI checks the SHA-256 and the signature before it writes a file.
Package0853729b…116c96d3
0853729b13dd…e53d116c96d31 file · ⁦1.1 KB⁩
ScanLow1 finding
  • Links to outside domains
See the full scan
Sourcegithub.com/AeonDave/malskill/tree/main/offensive-tools/fuzzing/radamsa@21cd568

Mutation-focused fuzzer/generator: excellent as an input transformer in pipelines.

Quick Start

bash
# Example mutation from file -> stdout
radamsa seed.bin > mutated.bin

# Batch generate samples
radamsa -n 100 -o out-%n.bin seeds/*

Operator Flow

  1. Collect representative seed samples (multiple real-world variants).
  2. Generate mutations in deterministic mode when triaging (--seed).
  3. Pipe outputs into target/harness and monitor crash/timeout/resource metrics.
  4. Preserve crashing case + seed metadata.
  5. Replay with identical seed/options before root-cause analysis.

Best Pattern

  • Use Radamsa as a feeder into:
    • web fuzzers,
    • API clients,
    • parser harnesses,
    • protocol replay tools.
  • Keep reproducibility via controlled seeds when triaging (--seed patterns).

Practical Tricks

  • Use %n output patterns with -o to preserve one-file-per-testcase workflows.
  • Prefer many diverse seeds over one huge synthetic seed.
  • For network workflows, pair with replay harnesses or server/client output modes supported by Radamsa.
  • Start broad first, then constrain mutator options only if throughput or relevance requires it.

Common Pitfalls

  • Treating Radamsa as coverage-guided: it is mutation-focused, not coverage-native.
  • Ignoring seed quality; weak seeds produce weak campaigns.
  • Running unreproducible random campaigns without seed tracking.

Notes

  • GitHub mirror is stale; upstream moved to GitLab.
  • Strong complement to coverage-guided engines (AFL++/libFuzzer/honggfuzz).

Resources

Files in this version

Permissions

PermissionDeclaredFound by the scan
Reads filesNoNo
Writes filesYesYes
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, 202621cd5680853729bSigned–

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