PYSEC-2026-2028
Vulnerability from pysec - Published: 2026-07-07 16:02 - Updated: 2026-07-07 17:25Impact
the slice() builtin can elide side effects when the output length is 0, and the source bytestring is a builtin (msg.data or <address>.code). the reason is that for these source locations, the check that length >= 1 is skipped:
https://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/builtins/functions.py#L315-L319
the result is that a 0-length bytestring constructed with slice can be passed to make_byte_array_copier, which elides evaluation of its source argument when the max length is 0:
https://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/codegen/core.py#L189-L191
the impact is that side effects in the start argument may be elided when the length argument is 0, e.g. slice(msg.data, self.do_side_effect(), 0).
the following example illustrates how the issue would look in user code
counter: public(uint256)
@external
def test() -> Bytes[10]:
b: Bytes[10] = slice(msg.data, self.side_effect(), 0)
return b
def side_effect() -> uint256:
self.counter += 1
return 0
the severity assigned is low, since this is not a very useful pattern and unlikely to be found in user code.
Patches
the fix is tracked in https://github.com/vyperlang/vyper/pull/4645, which disallows any invocation of slice() with length 0, including for the ad hoc locations discussed in this advisory.
Workarounds
Is there a way for users to fix or remediate the vulnerability without upgrading?
References
Are there any links users can visit to find out more?
| Name | purl | vyper | pkg:pypi/vyper |
|---|
{
"affected": [
{
"package": {
"ecosystem": "PyPI",
"name": "vyper",
"purl": "pkg:pypi/vyper"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"last_affected": "0.4.2rc1"
}
],
"type": "ECOSYSTEM"
}
],
"versions": [
"0.1.0b1",
"0.1.0b10",
"0.1.0b11",
"0.1.0b12",
"0.1.0b13",
"0.1.0b14",
"0.1.0b15",
"0.1.0b16",
"0.1.0b17",
"0.1.0b2",
"0.1.0b3",
"0.1.0b4",
"0.1.0b5",
"0.1.0b6",
"0.1.0b7",
"0.1.0b8",
"0.1.0b9",
"0.2.1",
"0.2.10",
"0.2.11",
"0.2.12",
"0.2.13",
"0.2.14",
"0.2.15",
"0.2.16",
"0.2.2",
"0.2.3",
"0.2.4",
"0.2.5",
"0.2.6",
"0.2.7",
"0.2.8",
"0.2.9",
"0.3.0",
"0.3.1",
"0.3.10",
"0.3.10rc1",
"0.3.10rc2",
"0.3.10rc3",
"0.3.10rc4",
"0.3.10rc5",
"0.3.2",
"0.3.3",
"0.3.4",
"0.3.5",
"0.3.6",
"0.3.7",
"0.3.8",
"0.3.9",
"0.4.0",
"0.4.0b1",
"0.4.0b2",
"0.4.0b3",
"0.4.0b4",
"0.4.0b5",
"0.4.0b6",
"0.4.0rc1",
"0.4.0rc2",
"0.4.0rc3",
"0.4.0rc4",
"0.4.0rc5",
"0.4.0rc6",
"0.4.1",
"0.4.1b1",
"0.4.1b2",
"0.4.1b3",
"0.4.1b4",
"0.4.1rc1",
"0.4.1rc2",
"0.4.1rc3",
"0.4.2rc1"
]
}
],
"aliases": [
"CVE-2025-47774",
"GHSA-3vcg-j39x-cwfm"
],
"details": "### Impact\nthe `slice()` builtin can elide side effects when the output length is 0, and the source bytestring is a builtin (`msg.data` or `\u003caddress\u003e.code`). the reason is that for these source locations, the check that `length \u003e= 1` is skipped:\nhttps://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/builtins/functions.py#L315-L319\n\nthe result is that a 0-length bytestring constructed with slice can be passed to `make_byte_array_copier`, which elides evaluation of its source argument when the max length is 0:\nhttps://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/codegen/core.py#L189-L191\n\nthe impact is that side effects in the `start` argument may be elided when the `length` argument is 0, e.g. `slice(msg.data, self.do_side_effect(), 0)`.\n\nthe following example illustrates how the issue would look in user code\n```vyper\ncounter: public(uint256)\n\n@external\ndef test() -\u003e Bytes[10]:\n b: Bytes[10] = slice(msg.data, self.side_effect(), 0)\n return b\n\ndef side_effect() -\u003e uint256:\n self.counter += 1\n return 0\n```\n\nthe severity assigned is low, since this is not a very useful pattern and unlikely to be found in user code.\n\n### Patches\n\nthe fix is tracked in https://github.com/vyperlang/vyper/pull/4645, which disallows any invocation of `slice()` with length 0, including for the ad hoc locations discussed in this advisory.\n\n### Workarounds\n_Is there a way for users to fix or remediate the vulnerability without upgrading?_\n\n### References\n_Are there any links users can visit to find out more?_",
"id": "PYSEC-2026-2028",
"modified": "2026-07-07T17:25:48.713294Z",
"published": "2026-07-07T16:02:52.607813Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/vyperlang/vyper/security/advisories/GHSA-3vcg-j39x-cwfm"
},
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2025-47774"
},
{
"type": "WEB",
"url": "https://github.com/vyperlang/vyper/pull/4645"
},
{
"type": "PACKAGE",
"url": "https://github.com/vyperlang/vyper"
},
{
"type": "WEB",
"url": "https://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/builtins/functions.py#L315-L319"
},
{
"type": "WEB",
"url": "https://github.com/vyperlang/vyper/blob/68b68c4b30c5ef2f312b4674676170b8a6eaa316/vyper/codegen/core.py#L189-L191"
},
{
"type": "PACKAGE",
"url": "https://pypi.org/project/vyper"
},
{
"type": "ADVISORY",
"url": "https://github.com/advisories/GHSA-3vcg-j39x-cwfm"
}
],
"severity": [
{
"score": "CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:N/VC:N/VI:L/VA:N/SC:N/SI:N/SA:N/E:P",
"type": "CVSS_V4"
}
],
"summary": "Vyper\u0027s `slice()` may elide side-effects when output length is 0"
}
Drag nodes to arrange them. Double-click a related vulnerability to open it, or a source to visit its link. The Physics button on the left tunes the layout and the simulation; the legend at the bottom left hides or shows nodes by kind and by sighting type.
Sightings
| Author | Source | Type | Date | Other |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or observed by the user.
- Confirmed: The vulnerability has been validated from an analyst's perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: The vulnerability was observed as exploited by the user who reported the sighting.
- Patched: The vulnerability was observed as successfully patched by the user who reported the sighting.
- Not exploited: The vulnerability was not observed as exploited by the user who reported the sighting.
- Not confirmed: The user expressed doubt about the validity of the vulnerability.
- Not patched: The vulnerability was not observed as successfully patched by the user who reported the sighting.
Vulnerability timeline
Sightings and recorded dates, grouped by day · UTC
How durations are calculated
Following the VTEM framework (Black Hat USA 2015), time to patch starts when the vendor or a coordinator was informed, total time to patch at discovery, and disclosure to patch at public disclosure; vendor response time runs from the report to its acknowledgement. Time to exploit and time unpatched start at the earliest discovery or report, falling back to public disclosure. Unpatched time ends at the first recorded patch, or the calculation time when no patch is recorded. Exploitation to patch shows the interval between those two events; a published proof of concept is a public exploit, not exploitation.
Each result shows its source dates. A negative interval means the end event came first; when date-only inputs leave the order open, the result says so. Date-only inputs are approximate. The earliest evidence wins: KEV report dates and sightings are labelled as such. Records describing another vulnerability are listed but never calculated from, and a PoC alone does not establish exploitation. Missing patch evidence does not prove that no patch exists. Filters below change the event list only.
Loading timeline…
The approach is described in our paper Mapping CVEs to MITRE ATT&CK Techniques: A Curated Gold-Set Classifier and the Limits of LLM-Assisted Label Expansion.
Browse all ATT&CK techniques and the vulnerabilities related to each.
Related by attack behaviour
Vulnerabilities whose description is nearest to this one in the vector space of the CIRCL/vulnerability-attack-technique-biencoder model. This is a similarity search over the bi-encoder space (plain cosine), not a classification, and it has no measured accuracy.