CVE-2026-61536 (GCVE-0-2026-61536)
Vulnerability from cvelistv5 – Published: 2026-07-30 16:50 – Updated: 2026-07-31 23:01
VLAI
EPSS
VEX
Title
Banks: Unsafe importlib.import_module of attacker-controlled Tool.import_path in CompletionExtension allows RCE
Summary
Banks generates meaningful LLM prompts using a simple template language. In versions prior to 2.4.3, banks parses Tool JSON objects from the rendered body of {% completion %} blocks and later resolves their import_path field through importlib.import_module(...) + getattr(...) to obtain the callable that handles a tool call. There is no allowlist or sanitization on import_path, so any importable Python attribute (e.g. os.system, subprocess.getoutput) can be selected. When the LLM emits a tool_calls entry whose function.name matches the attacker-supplied tool name, the resolved callable is invoked with kwargs decoded from tool_call.function.arguments, yielding arbitrary code execution in the banks-hosting process. This is distinct from GHSA-gphh-9q3h-jgpp / CVE-2026-44209. That advisory was fixed in 2.4.2 by switching src/banks/env.py from Environment to SandboxedEnvironment. The fix does not touch src/banks/extensions/completion.py, and the unsafe import + getattr chain still executes on 2.4.2. The malicious Tool JSON is plain text in the rendered template body — it requires no Jinja attribute access, so the sandbox is irrelevant. This issue has been fixed in version 2.4.3.
Severity
7.5 (High)
SSVC
Exploitation: poc
Automatable: no
Technical Impact: total
CISA Coordinator (v2.0.3)
CWE
Assigner
References
1 reference
| URL | Tags |
|---|---|
| https://github.com/masci/banks/security/advisorie… | x_refsource_CONFIRM |
{
"containers": {
"adp": [
{
"metrics": [
{
"other": {
"content": {
"id": "CVE-2026-61536",
"options": [
{
"Exploitation": "poc"
},
{
"Automatable": "no"
},
{
"Technical Impact": "total"
}
],
"role": "CISA Coordinator",
"timestamp": "2026-07-31T23:00:33.742984Z",
"version": "2.0.3"
},
"type": "ssvc"
}
}
],
"providerMetadata": {
"dateUpdated": "2026-07-31T23:01:03.395Z",
"orgId": "134c704f-9b21-4f2e-91b3-4a467353bcc0",
"shortName": "CISA-ADP"
},
"references": [
{
"tags": [
"exploit"
],
"url": "https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7"
}
],
"title": "CISA ADP Vulnrichment"
}
],
"cna": {
"affected": [
{
"product": "banks",
"vendor": "masci",
"versions": [
{
"status": "affected",
"version": "\u003c 2.4.3"
}
]
}
],
"descriptions": [
{
"lang": "en",
"value": "Banks generates meaningful LLM prompts using a simple template language. In versions prior to 2.4.3, banks parses Tool JSON objects from the rendered body of {% completion %} blocks and later resolves their import_path field through importlib.import_module(...) + getattr(...) to obtain the callable that handles a tool call. There is no allowlist or sanitization on import_path, so any importable Python attribute (e.g. os.system, subprocess.getoutput) can be selected. When the LLM emits a tool_calls entry whose function.name matches the attacker-supplied tool name, the resolved callable is invoked with kwargs decoded from tool_call.function.arguments, yielding arbitrary code execution in the banks-hosting process. This is distinct from GHSA-gphh-9q3h-jgpp / CVE-2026-44209. That advisory was fixed in 2.4.2 by switching src/banks/env.py from Environment to SandboxedEnvironment. The fix does not touch src/banks/extensions/completion.py, and the unsafe import + getattr chain still executes on 2.4.2. The malicious Tool JSON is plain text in the rendered template body \u2014 it requires no Jinja attribute access, so the sandbox is irrelevant. This issue has been fixed in version 2.4.3."
}
],
"metrics": [
{
"cvssV3_1": {
"attackComplexity": "HIGH",
"attackVector": "NETWORK",
"availabilityImpact": "HIGH",
"baseScore": 7.5,
"baseSeverity": "HIGH",
"confidentialityImpact": "HIGH",
"integrityImpact": "HIGH",
"privilegesRequired": "LOW",
"scope": "UNCHANGED",
"userInteraction": "NONE",
"vectorString": "CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H",
"version": "3.1"
}
}
],
"problemTypes": [
{
"descriptions": [
{
"cweId": "CWE-94",
"description": "CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)",
"lang": "en",
"type": "CWE"
}
]
},
{
"descriptions": [
{
"cweId": "CWE-470",
"description": "CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)",
"lang": "en",
"type": "CWE"
}
]
}
],
"providerMetadata": {
"dateUpdated": "2026-07-30T16:50:01.730Z",
"orgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"shortName": "GitHub_M"
},
"references": [
{
"name": "https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7",
"tags": [
"x_refsource_CONFIRM"
],
"url": "https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7"
}
],
"source": {
"advisory": "GHSA-64vx-6h2c-rjh7",
"discovery": "UNKNOWN"
},
"title": "Banks: Unsafe importlib.import_module of attacker-controlled Tool.import_path in CompletionExtension allows RCE"
}
},
"cveMetadata": {
"assignerOrgId": "a0819718-46f1-4df5-94e2-005712e83aaa",
"assignerShortName": "GitHub_M",
"cveId": "CVE-2026-61536",
"datePublished": "2026-07-30T16:50:01.730Z",
"dateReserved": "2026-07-10T16:27:03.093Z",
"dateUpdated": "2026-07-31T23:01:03.395Z",
"state": "PUBLISHED"
},
"dataType": "CVE_RECORD",
"dataVersion": "5.2",
"vulnerability-lookup:meta": {
"epss": {
"cve": "CVE-2026-61536",
"date": "2026-07-31",
"epss": "0.003",
"percentile": "0.22308"
},
"nvd": "{\"cve\":{\"id\":\"CVE-2026-61536\",\"sourceIdentifier\":\"security-advisories@github.com\",\"published\":\"2026-07-30T19:18:34.553\",\"lastModified\":\"2026-07-31T23:17:25.930\",\"vulnStatus\":\"Deferred\",\"cveTags\":[],\"descriptions\":[{\"lang\":\"en\",\"value\":\"Banks generates meaningful LLM prompts using a simple template language. In versions prior to 2.4.3, banks parses Tool JSON objects from the rendered body of {% completion %} blocks and later resolves their import_path field through importlib.import_module(...) + getattr(...) to obtain the callable that handles a tool call. There is no allowlist or sanitization on import_path, so any importable Python attribute (e.g. os.system, subprocess.getoutput) can be selected. When the LLM emits a tool_calls entry whose function.name matches the attacker-supplied tool name, the resolved callable is invoked with kwargs decoded from tool_call.function.arguments, yielding arbitrary code execution in the banks-hosting process. This is distinct from GHSA-gphh-9q3h-jgpp / CVE-2026-44209. That advisory was fixed in 2.4.2 by switching src/banks/env.py from Environment to SandboxedEnvironment. The fix does not touch src/banks/extensions/completion.py, and the unsafe import + getattr chain still executes on 2.4.2. The malicious Tool JSON is plain text in the rendered template body \u2014 it requires no Jinja attribute access, so the sandbox is irrelevant. This issue has been fixed in version 2.4.3.\"}],\"affected\":[{\"source\":\"security-advisories@github.com\",\"affectedData\":[{\"vendor\":\"masci\",\"product\":\"banks\",\"versions\":[{\"version\":\"\u003c 2.4.3\",\"status\":\"affected\"}]}]}],\"metrics\":{\"cvssMetricV31\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"cvssData\":{\"version\":\"3.1\",\"vectorString\":\"CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H\",\"baseScore\":7.5,\"baseSeverity\":\"HIGH\",\"attackVector\":\"NETWORK\",\"attackComplexity\":\"HIGH\",\"privilegesRequired\":\"LOW\",\"userInteraction\":\"NONE\",\"scope\":\"UNCHANGED\",\"confidentialityImpact\":\"HIGH\",\"integrityImpact\":\"HIGH\",\"availabilityImpact\":\"HIGH\"},\"exploitabilityScore\":1.6,\"impactScore\":5.9}],\"ssvcV203\":[{\"source\":\"134c704f-9b21-4f2e-91b3-4a467353bcc0\",\"ssvcData\":{\"timestamp\":\"2026-07-31T23:00:33.742984Z\",\"id\":\"CVE-2026-61536\",\"options\":[{\"exploitation\":\"poc\"},{\"automatable\":\"no\"},{\"technicalImpact\":\"total\"}],\"role\":\"CISA Coordinator\",\"version\":\"2.0.3\"}}]},\"weaknesses\":[{\"source\":\"security-advisories@github.com\",\"type\":\"Secondary\",\"description\":[{\"lang\":\"en\",\"value\":\"CWE-94\"},{\"lang\":\"en\",\"value\":\"CWE-470\"}]}],\"references\":[{\"url\":\"https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7\",\"source\":\"security-advisories@github.com\"},{\"url\":\"https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7\",\"source\":\"134c704f-9b21-4f2e-91b3-4a467353bcc0\"}]}}",
"vulnrichment": {
"containers": "{\"adp\": [{\"title\": \"CISA ADP Vulnrichment\", \"metrics\": [{\"other\": {\"type\": \"ssvc\", \"content\": {\"id\": \"CVE-2026-61536\", \"role\": \"CISA Coordinator\", \"options\": [{\"Exploitation\": \"poc\"}, {\"Automatable\": \"no\"}, {\"Technical Impact\": \"total\"}], \"version\": \"2.0.3\", \"timestamp\": \"2026-07-31T23:00:33.742984Z\"}}}], \"references\": [{\"url\": \"https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7\", \"tags\": [\"exploit\"]}], \"providerMetadata\": {\"orgId\": \"134c704f-9b21-4f2e-91b3-4a467353bcc0\", \"shortName\": \"CISA-ADP\", \"dateUpdated\": \"2026-07-31T23:00:59.036Z\"}}], \"cna\": {\"title\": \"Banks: Unsafe importlib.import_module of attacker-controlled Tool.import_path in CompletionExtension allows RCE\", \"source\": {\"advisory\": \"GHSA-64vx-6h2c-rjh7\", \"discovery\": \"UNKNOWN\"}, \"metrics\": [{\"cvssV3_1\": {\"scope\": \"UNCHANGED\", \"version\": \"3.1\", \"baseScore\": 7.5, \"attackVector\": \"NETWORK\", \"baseSeverity\": \"HIGH\", \"vectorString\": \"CVSS:3.1/AV:N/AC:H/PR:L/UI:N/S:U/C:H/I:H/A:H\", \"integrityImpact\": \"HIGH\", \"userInteraction\": \"NONE\", \"attackComplexity\": \"HIGH\", \"availabilityImpact\": \"HIGH\", \"privilegesRequired\": \"LOW\", \"confidentialityImpact\": \"HIGH\"}}], \"affected\": [{\"vendor\": \"masci\", \"product\": \"banks\", \"versions\": [{\"status\": \"affected\", \"version\": \"\u003c 2.4.3\"}]}], \"references\": [{\"url\": \"https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7\", \"name\": \"https://github.com/masci/banks/security/advisories/GHSA-64vx-6h2c-rjh7\", \"tags\": [\"x_refsource_CONFIRM\"]}], \"descriptions\": [{\"lang\": \"en\", \"value\": \"Banks generates meaningful LLM prompts using a simple template language. In versions prior to 2.4.3, banks parses Tool JSON objects from the rendered body of {% completion %} blocks and later resolves their import_path field through importlib.import_module(...) + getattr(...) to obtain the callable that handles a tool call. There is no allowlist or sanitization on import_path, so any importable Python attribute (e.g. os.system, subprocess.getoutput) can be selected. When the LLM emits a tool_calls entry whose function.name matches the attacker-supplied tool name, the resolved callable is invoked with kwargs decoded from tool_call.function.arguments, yielding arbitrary code execution in the banks-hosting process. This is distinct from GHSA-gphh-9q3h-jgpp / CVE-2026-44209. That advisory was fixed in 2.4.2 by switching src/banks/env.py from Environment to SandboxedEnvironment. The fix does not touch src/banks/extensions/completion.py, and the unsafe import + getattr chain still executes on 2.4.2. The malicious Tool JSON is plain text in the rendered template body \\u2014 it requires no Jinja attribute access, so the sandbox is irrelevant. This issue has been fixed in version 2.4.3.\"}], \"problemTypes\": [{\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-94\", \"description\": \"CWE-94: Improper Control of Generation of Code (\u0027Code Injection\u0027)\"}]}, {\"descriptions\": [{\"lang\": \"en\", \"type\": \"CWE\", \"cweId\": \"CWE-470\", \"description\": \"CWE-470: Use of Externally-Controlled Input to Select Classes or Code (\u0027Unsafe Reflection\u0027)\"}]}], \"providerMetadata\": {\"orgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"shortName\": \"GitHub_M\", \"dateUpdated\": \"2026-07-30T16:50:01.730Z\"}}}",
"cveMetadata": "{\"cveId\": \"CVE-2026-61536\", \"state\": \"PUBLISHED\", \"dateUpdated\": \"2026-07-31T23:01:03.395Z\", \"dateReserved\": \"2026-07-10T16:27:03.093Z\", \"assignerOrgId\": \"a0819718-46f1-4df5-94e2-005712e83aaa\", \"datePublished\": \"2026-07-30T16:50:01.730Z\", \"assignerShortName\": \"GitHub_M\"}",
"dataType": "CVE_RECORD",
"dataVersion": "5.2"
}
}
}
Loading…
Loading…
Experimental. This forecast is provided for visualization only and may change without notice. Do not use it for operational decisions.
Forecast uses a logistic model when the trend is rising, or an exponential decay model when the trend is falling. Fitted via linearized least squares.
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.
Loading…
The MITRE ATT&CK techniques below are AI-generated suggestions, inferred from the description of the
vulnerability by the CIRCL/vulnerability-attack-technique-classification-roberta-base
model, served locally by ML-Gateway.
They have not been verified by an analyst and are provided for guidance only.
Loading…
Loading…