ai-orchestration-langchain

Warn

Audited by Socket on Sep 20, 2026

2 alerts found:

Securityx2
SecurityMEDIUM
examples/structured-output-tools.md

The fragment is legitimate educational documentation and shows no clear supply-chain malware. However, the `calculator` example contains a high-impact arbitrary-code-execution vulnerability because it evaluates an untrusted expression with `eval`. It should be replaced with a restricted math parser or isolated sandbox before use with external or model-controlled input. External model transmission and logging should also be reviewed for data-sensitivity concerns.

Confidence: 99%Severity: 78%
SecurityMEDIUM
examples/agents.md

The fragment is benign documentation with one significant security issue: the calculator tool evaluates untrusted string input using `eval`, creating a potential arbitrary JavaScript execution vulnerability when exposed to users or model-generated tool calls. No evidence of supply-chain malware or intentional malicious behavior is present.

Confidence: 99%Severity: 72%
Audit Metadata
Analyzed At
Sep 20, 2026, 03:57 PM
Package URL
pkg:socket/skills-sh/agents-inc%2Fskills%2Fai-orchestration-langchain%2F@2965dc48431c8e16570dd81199a14ed3c48100babd130ebeebbc1a8e9adbe092
Security Audit — socket — ai-orchestration-langchain