DIGGINGBEAGLE

Case · DiggingBeagle record

EchoFuzz used LLM-guided fuzzing to reach deeper smart-contract states

ICSE 2026 research introduced EchoFuzz, an LLM-guided Ethereum smart-contract fuzzer that uses contract logic, static analysis and runtime coverage feedback to generate and refine Vulnerable Function Call Sequences. The authors report materially higher coverage and vulnerability detection than the compared fuzzers and 37 previously unknown vulnerabilities across 19 real-world contract projects. Those 37 findings remain researcher-reported rather than independently confirmed zero-days; the public disclosure repository says most teams could not be contacted and CVE applications are still in progress.

Scope

Defensive/research capability result concerning LLM-assisted Ethereum smart-contract vulnerability discovery. This Case is not a confirmed live exploitation incident, autonomous fund theft, governance manipulation, or evidence that any of the 37 reported findings were exploited on-chain.

Assessment profile

Assessment method
Setting
research demonstration
Exploitation
demonstrated
Evidence
Grade B
Basis and provenance

Detailed method, benchmark, repository and disclosure claims support a research capability result. The 37 researcher-listed findings, disputed D3 count and unresolved maintainer confirmation do not establish mainnet exploitation or an aggregate fix. Assessment is a desk review of the retained canonical Claims and cited Source metadata at their recorded cutoff, not a new external verification. Impact remains ungraded.

Assessed Sep 24, 2026 using diggingbeagle.assessment/1.

Evidence basis: CLM-ECHOFUZZ-METHOD · CLM-ECHOFUZZ-37-VULNS · CLM-ECHOFUZZ-DISCLOSURE-STATUS · CLM-ECHOFUZZ-REALWORLD-DATASET · EchoFuzz: Empowering Smart Contract Fuzzing with Large Language Models · EchoFuzz real-world vulnerability disclosure dataset

At a glance

Timeline

  1. Apr 16, 2026
    Event type unspecified

    EchoFuzz presented in the ICSE 2026 Research Track

    The official ICSE program places EchoFuzz in the Testing and Analysis 8 session at 12:00-12:15.

  2. Sep 13, 2026
    Event type unspecified

    EchoFuzz arXiv version posted

    arXiv v1 was submitted with ICSE 2026 conference metadata and DOI 10.1145/3744916.3773166.

  3. Sep 13, 2026
    disclosure

    Public vulnerability artifact accompanies the research

    The researcher repository enumerates 19 project/address entries and 37 reported findings, states that most teams could not be contacted, and says CVE applications are in progress.

Claims & evidence

CLM-ECHOFUZZ-METHODEchoFuzz combines LLM reasoning, Slither-assisted static analysis and runtime fuzzing feedback to generate candidate Vulnerable Function Call Sequences and iteratively redirect fuzzing toward uncovered branches.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-37-VULNSThe EchoFuzz authors report 37 previously unknown vulnerabilities across 19 projects in the real-world dataset; the public researcher repository enumerates 19 project/address entries totaling 37 reported vulnerability instances.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-OVERHEADThe reported average time to test one contract is 838 seconds for EchoFuzz versus 603 seconds for MuFuzz; the authors attribute 28% of EchoFuzz's total execution time to LLM overhead and report 376 detected vulnerabilities versus 231 for MuFuzz in this comparison.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-BENCHMARKThe paper reports 62.77% more detected vulnerabilities than MuFuzz on average across D1-D3. In the detailed D1 branch-coverage comparison, EchoFuzz reaches 79.1% on medium contracts (+29.19% over MuFuzz) and 76.70% on large contracts (+32.24% over MuFuzz); the paper also summarizes branch-coverage improvement as 29.19% in its contributions text.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-ITERATIONThe paper's ablation study attributes substantial additional coverage and vulnerability discovery to LLM-guided sequence generation and runtime feedback: the complete framework reports 10.65% more coverage than the no-iteration variant at 300 seconds and 54.17% more vulnerabilities on D3.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-CONFERENCEEchoFuzz appears in the ICSE 2026 Research Track program, with a presentation scheduled for April 16, 2026; the arXiv version identifies DOI 10.1145/3744916.3773166 for the conference publication.supported

Basis: direct observation

Link to claim
CLM-ECHOFUZZ-FUNDRAISERIn the paper's motivating Fundraiser example, sFuzz, Confuzzius, IR-Fuzz and MuFuzz miss the target bug, while EchoFuzz derives the sequence invest -> invest -> withdraw to trigger it; a later feedback-guided sequence invest -> refund raises reported branch coverage to 100% in the case study.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-LIMITATIONSThe paper reports that EchoFuzz can still produce false negatives, incurs greater execution time than MuFuzz in the reported setup, and becomes less reliable as contract size exceeds the LLM context capacity.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-OPEN-SOURCEThe EchoFuzz implementation is publicly available with LLM, dataset, fuzzer, prompt and orchestration components, and the repository links the ICSE 2026 paper.supported

Basis: direct observation

  • supports
    EchoFuzz public implementationsource code

    Repository root and README, including LLM, dataset, fuzzer, prompts, chain_guided.py and iteration_process.py

Link to claim
CLM-ECHOFUZZ-FALSE-NEGATIVESEchoFuzz has documented false negatives. The paper attributes two gasless false negatives relative to MuFuzz to insufficient cumulative seed iteration and exploration; an experimental EchoFuzz+LTL variant with manually generated temporal properties resolves both in the reported test.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-DISCLOSURE-STATUSThe researchers' public vulnerability repository states that most affected contract teams could not be directly contacted and that CVE applications are in progress. The checked public evidence therefore does not independently confirm all 37 findings as production zero-days.supported

Basis: direct observation

Link to claim
CLM-ECHOFUZZ-MODEL-SENSITIVITYEchoFuzz performance depends materially on the selected LLM. In D1 VFCS generation, Claude 3.5 Sonnet reaches a 94.2% usable-VFCS ratio, while LLaMA 3.1 8B reaches 42.1%; on large contracts several tested models fall below the IR-Fuzz branch-coverage baseline.supported

Basis: reported finding

Link to claim
CLM-ECHOFUZZ-REALWORLD-DATASETSection 5.1.2 describes D3 as 146 large and complex contract projects manually scraped from Etherscan, and the disclosure repository also says 146; however, Table 1 of the same paper labels D3 as 145, creating a one-project internal count discrepancy.supported

Basis: direct observation

Link to claim
CLM-ECHOFUZZ-VULNERABILITY-MIXThe 37 researcher-listed findings break down to 15 integer over/under-flow, 8 gasless, 7 reentrancy, 3 timestamp dependency, 2 unexpected Ether/Ether frozen, 1 dangerous delegatecall and 1 unchecked-call finding; no block-number-dependency finding appears in the listed 19 projects.supported

Basis: direct observation

Link to claim

Implications

EchoFuzz provides peer-reviewed evidence that LLM reasoning can improve the selection of state-transition sequences used by smart-contract fuzzers, particularly for larger contracts where useful vulnerable states may require non-obvious call ordering. For Digging Beagle it extends the existing KASS record backward in the attack lifecycle: AI assistance is being measured not only in exploit construction after a vulnerability is known, but also in the search process that discovers candidate vulnerabilities.

Controls and mitigations

  • Treat EchoFuzz as an additional discovery layer, not a replacement for static analysis, invariant testing, conventional fuzzing, manual audit or exploit validation.
  • Independently reproduce high-impact findings with deterministic tests and preserve the exact contract version, transaction sequence, state assumptions and vulnerability oracle.
  • Run AI-guided fuzzing in isolated local or forked environments without production signing authority.
  • Use multiple analysis strategies: the paper itself documents cases where MuFuzz finds issues EchoFuzz misses.
  • For large contracts, use segmentation or hierarchical analysis because the paper reports declining LLM response success as contract size exceeds context capacity.
  • Consider temporal/access-control properties or equivalent state-machine constraints when sequence diversity is insufficient; the paper's manual-LTL experiment resolved two documented false negatives.
  • Do not label researcher-reported findings as confirmed zero-days until maintainer acknowledgement, independent reproduction, CVE assignment or another strong confirmation exists.

Unknowns and contradictions

  • Whether any of the 37 researcher-reported findings have since been independently reproduced or acknowledged by the affected project maintainers.
  • Whether CVE identifiers have now been assigned to any of the 37 findings.
  • Whether any affected contract remained economically active or held material value at the time of testing or disclosure.
  • Whether the D3 count is 145 or 146; the paper contains both numbers.
  • Whether the 37 findings are all unique root causes rather than multiple oracle classifications over related code paths; the public artifact lists categories and addresses but not full independent triage for each finding.
  • Whether later LLMs materially change the model-sensitivity results; the paper evaluates models including Claude 3.5 Sonnet, GPT-4o-mini, GPT-3.5 Turbo and LLaMA/Qwen variants.
  • No evidence in the checked sources establishes autonomous mainnet exploitation, theft, governance manipulation or malicious use of EchoFuzz.

Sources and citation

EvidenceEchoFuzz: Empowering Smart Contract Fuzzing with Large Language ModelsEvidenceEchoFuzz real-world vulnerability disclosure datasetEvidenceEchoFuzz - ICSE 2026 Research TrackEvidenceEchoFuzz public implementation

Material revision history

  1. Sep 21, 2026 · Published version · first publication · revision 31
  2. Oct 5, 2026 · Published version · material revision · revision 37Public fields changed: assessment_profile
Record metadata

Canonical ID: CASE-2026-ECHOFUZZ · revision 37