Vulnerabilites related to abetlen - llama-cpp-python
CVE-2024-34359 (GCVE-0-2024-34359)
Vulnerability from cvelistv5
Published
2024-05-10 17:07
Modified
2024-08-02 02:51
Severity ?
VLAI Severity ?
EPSS score ?
CWE
- CWE-76 - Improper Neutralization of Equivalent Special Elements
Summary
llama-cpp-python is the Python bindings for llama.cpp. `llama-cpp-python` depends on class `Llama` in `llama.py` to load `.gguf` llama.cpp or Latency Machine Learning Models. The `__init__` constructor built in the `Llama` takes several parameters to configure the loading and running of the model. Other than `NUMA, LoRa settings`, `loading tokenizers,` and `hardware settings`, `__init__` also loads the `chat template` from targeted `.gguf` 's Metadata and furtherly parses it to `llama_chat_format.Jinja2ChatFormatter.to_chat_handler()` to construct the `self.chat_handler` for this model. Nevertheless, `Jinja2ChatFormatter` parse the `chat template` within the Metadate with sandbox-less `jinja2.Environment`, which is furthermore rendered in `__call__` to construct the `prompt` of interaction. This allows `jinja2` Server Side Template Injection which leads to remote code execution by a carefully constructed payload.
References
► | URL | Tags |
---|---|---|
Impacted products
Vendor | Product | Version | ||
---|---|---|---|---|
abetlen | llama-cpp-python |
Version: >= 0.2.30, <= 0.2.71 |
{ "containers": { "adp": [ { "affected": [ { "cpes": [ "cpe:2.3:a:abetlen:llama-cpp-python:*:*:*:*:*:*:*:*" ], "defaultStatus": "unknown", "product": "llama-cpp-python", "vendor": "abetlen", "versions": [ { "lessThanOrEqual": "0.2.71", "status": "affected", "version": "0.2.30", "versionType": "custom" } ] } ], "metrics": [ { "other": { "content": { "id": "CVE-2024-34359", "options": [ { "Exploitation": "poc" }, { "Automatable": "no" }, { "Technical Impact": "total" } ], "role": "CISA Coordinator", "timestamp": "2024-05-15T19:35:24.408358Z", "version": "2.0.3" }, "type": "ssvc" } } ], "providerMetadata": { "dateUpdated": "2024-06-06T18:29:15.313Z", "orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0", "shortName": "CISA-ADP" }, "title": "CISA ADP Vulnrichment" }, { "providerMetadata": { "dateUpdated": "2024-08-02T02:51:10.739Z", "orgId": "af854a3a-2127-422b-91ae-364da2661108", "shortName": "CVE" }, "references": [ { "name": "https://github.com/abetlen/llama-cpp-python/security/advisories/GHSA-56xg-wfcc-g829", "tags": [ "x_refsource_CONFIRM", "x_transferred" ], "url": "https://github.com/abetlen/llama-cpp-python/security/advisories/GHSA-56xg-wfcc-g829" }, { "name": "https://github.com/abetlen/llama-cpp-python/commit/b454f40a9a1787b2b5659cd2cb00819d983185df", "tags": [ "x_refsource_MISC", "x_transferred" ], "url": "https://github.com/abetlen/llama-cpp-python/commit/b454f40a9a1787b2b5659cd2cb00819d983185df" } ], "title": "CVE Program Container" } ], "cna": { "affected": [ { "product": "llama-cpp-python", "vendor": "abetlen", "versions": [ { "status": "affected", "version": "\u003e= 0.2.30, \u003c= 0.2.71" } ] } ], "descriptions": [ { "lang": "en", "value": "llama-cpp-python is the Python bindings for llama.cpp. `llama-cpp-python` depends on class `Llama` in `llama.py` to load `.gguf` llama.cpp or Latency Machine Learning Models. The `__init__` constructor built in the `Llama` takes several parameters to configure the loading and running of the model. Other than `NUMA, LoRa settings`, `loading tokenizers,` and `hardware settings`, `__init__` also loads the `chat template` from targeted `.gguf` \u0027s Metadata and furtherly parses it to `llama_chat_format.Jinja2ChatFormatter.to_chat_handler()` to construct the `self.chat_handler` for this model. Nevertheless, `Jinja2ChatFormatter` parse the `chat template` within the Metadate with sandbox-less `jinja2.Environment`, which is furthermore rendered in `__call__` to construct the `prompt` of interaction. This allows `jinja2` Server Side Template Injection which leads to remote code execution by a carefully constructed payload." } ], "metrics": [ { "cvssV3_1": { "attackComplexity": "LOW", "attackVector": "NETWORK", "availabilityImpact": "HIGH", "baseScore": 9.7, "baseSeverity": "CRITICAL", "confidentialityImpact": "HIGH", "integrityImpact": "HIGH", "privilegesRequired": "NONE", "scope": "CHANGED", "userInteraction": "REQUIRED", "vectorString": "CVSS:3.1/AV:N/AC:L/PR:N/UI:R/S:C/C:H/I:H/A:H", "version": "3.1" } } ], "problemTypes": [ { "descriptions": [ { "cweId": "CWE-76", "description": "CWE-76: Improper Neutralization of Equivalent Special Elements", "lang": "en", "type": "CWE" } ] } ], "providerMetadata": { "dateUpdated": "2024-05-10T17:07:18.850Z", "orgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "shortName": "GitHub_M" }, "references": [ { "name": "https://github.com/abetlen/llama-cpp-python/security/advisories/GHSA-56xg-wfcc-g829", "tags": [ "x_refsource_CONFIRM" ], "url": "https://github.com/abetlen/llama-cpp-python/security/advisories/GHSA-56xg-wfcc-g829" }, { "name": "https://github.com/abetlen/llama-cpp-python/commit/b454f40a9a1787b2b5659cd2cb00819d983185df", "tags": [ "x_refsource_MISC" ], "url": "https://github.com/abetlen/llama-cpp-python/commit/b454f40a9a1787b2b5659cd2cb00819d983185df" } ], "source": { "advisory": "GHSA-56xg-wfcc-g829", "discovery": "UNKNOWN" }, "title": "llama-cpp-python vulnerable to Remote Code Execution by Server-Side Template Injection in Model Metadata" } }, "cveMetadata": { "assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa", "assignerShortName": "GitHub_M", "cveId": "CVE-2024-34359", "datePublished": "2024-05-10T17:07:18.850Z", "dateReserved": "2024-05-02T06:36:32.439Z", "dateUpdated": "2024-08-02T02:51:10.739Z", "state": "PUBLISHED" }, "dataType": "CVE_RECORD", "dataVersion": "5.1" }