registry / @caeliq/llms / 1.0.68
@caeliq/llms@1.0.68
A universal LLM API transformation server
AI Security Review
scanned 2mo ago · by lpm-firewall-aiStatic flags arise from bundled provider SDKs, OAuth support, and a token-count worker. Source inspection found no lifecycle trigger or concrete behavior that steals data, persists, or executes attacker-controlled commands.
Static reason
High-risk behavior combination matched malicious policy.
Trigger
Consumer imports or starts the server and configures a provider.
Impact
Configured request content and credentials may be sent to the selected provider, as expected for an LLM gateway.
Mechanism
Configured LLM request routing and OAuth token refresh.
AI rationale
Static flags arise from bundled provider SDKs, OAuth support, and a token-count worker. Source inspection found no lifecycle trigger or concrete behavior that steals data, persists, or executes attacker-controlled commands.
Decision evidence
public snapshotAI called this Clean at 96.0% confidence as Benign with low false-positive risk.
Evidence for AI clean decision
- Both runtime bundles include a token-count worker, explaining the process-execution heuristic without a shell command.
Evidence against
- package.json has no lifecycle scripts or self-dependency.
- The entrypoints are an LLM API transformation server and only run when imported or started by the consumer.
- Reviewed OAuth and provider networking is used to refresh configured provider credentials or forward configured LLM requests.
- No covert exfiltration endpoint, install-time action, persistence, or destructive behavior was found.
Behavioral surface
ChildProcessCryptoDynamicRequireEnvironmentVarsFilesystemNetwork
HighEntropyStringsMinifiedUrlStrings
Source & flagged code
10 flaggeddist/esm/server.mjsView file
1var $P=Object.create;var Sh=Object.defineProperty;var qP=Object.getOwnPropertyDescriptor;var HP=Object.getOwnPropertyNames;var zP=Object.getPrototypeOf,WP=Object.prototype.hasOwnPr...
L2: -- Key to operate on
...
L58: `:case"\u2028":case"\u2029":return V(),"";case"\r":return V(),lr()===`
L59: `&&V(),"";case"1":case"2":case"3":case"4":case"5":case"6":case"7":case"8":case"9":throw We(V());case void 0:throw We(V())}return V()}function s1(){let t="",e=lr();if(!st.isHexDigit...
L60: `+s;E=A.join(R),S=`{
...
L64: `+s+E+`,
L65: `+m+"]"}return o.pop(),s=m,A}}});var Yh=J((m5,cw)=>{var c1=sw(),u1=aw(),l1={parse:c1,stringify:u1};cw.exports=l1});var OA=J((rJ,bM)=>{bM.exports={"0":"\u2080","1":"\u2081","2":"\u2...
L66: `)||r,code:s,status:o},e.data.error)}return Object.assign({message:r,code:s,stat
Critical
Credential Exfiltration
Source appears to send environment or credential material to an external endpoint.
dist/esm/server.mjsView on unpkg · L1111${g}
L112: `+await t.crypto.sha256DigestHex(h),w=await _$(t.crypto,t.securityCredentials.secretAccessKey,i,t.region,r),m=await za(t.crypto,w,y),_=`${mR} Credential=${t.securityCredentials.acc...
L113: To learn more about authentication and Google APIs, visit:
High
Child Process
Package source references child process execution.
dist/esm/server.mjsView on unpkg · L111Findings
2 Critical7 High4 Medium3 Low
CriticalCredential Exfiltrationdist/esm/server.mjs
CriticalTrigger Reachable Dangerous Capabilitydist/cjs/server.cjs
HighChild Processdist/esm/server.mjs
HighSame File Env Network Executiondist/esm/server.mjs
HighCommand Output Exfiltrationdist/esm/server.mjs
HighCloud Metadata Accessdist/esm/server.mjs
HighCopied Package Dependency Bridgedist/esm/server.mjs
HighTrigger Reachable Credential Exfiltrationdist/cjs/server.cjs
HighTrigger Reachable Command Output Exfiltrationdist/cjs/server.cjs
MediumDynamic Requiredist/cjs/server.cjs
MediumNetwork
Medium
Affected versions and remediation
This report applies to @caeliq/llms@1.0.68.
See all versions of @caeliq/llms for other recorded verdicts.
- Review the evidence and your use of @caeliq/llms@1.0.68 before allowing it. Restrict the permissions described in this report.
- Choose an independently verified alternative or release. This report does not establish that other versions are safe.
Evidence last updated: .
Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.