registry / AI-agent capability abuse / @wengine-ai/claude-code-router-next / 2.3.2411

@wengine-ai/claude-code-router-next@2.3.2411

CLI for Claude Code Router

AI Security Review

scanned 10d ago · by lpm-firewall-ai

The transformer introduces a system-level instruction that steers AI tool use, creating an agent capability risk when selected. Because the behavior is exposed as a configurable router feature and no install-time activation was found, warn rather than block.

Static reason

High-risk behavior combination matched malicious policy.; source matched previously finalized malicious package; routed for review; source fingerprint signature matched known malicious package; routed for review

Trigger

An operator selects and uses the `tooluse` transformer for a routed request that includes tools.

Impact

This can steer the routed AI toward invoking available tools and change how its response is presented.

Mechanism

The transformer adds a system-level instruction encouraging tool use, requires a tool choice, and adds a synthetic exit tool whose call is converted into response content.

AI rationale

Evidence

dist/cli.js

Decision evidence

public snapshot

AI called this Suspicious at 91.0% confidence as Dangerous Capability with medium false-positive risk.

Evidence for warning

The `tooluse` transformer adds a system-level reminder urging the routed AI to use available tools and forces tool choice when tools are present.
The CLI lists `tooluse` as a selectable transformer and registers a dedicated transformer for it, so this behavior depends on configuration.
The transformer intercepts its synthetic exit-tool call and converts its arguments into response content.
The CLI derives per-project router configuration paths from the current working directory.

Evidence against

The behavior is part of an optional router transformer; the inspected source does not show it activating during package installation.

Behavioral surface

Source

ChildProcessCryptoDynamicRequireEnvironmentVarsEvalFilesystemNetworkShell

Supply chain

HighEntropyStringsMinifiedPossibleObfuscationUrlStrings

Source & flagged code

18 flagged

dist/cli.jsView file

8`:case"\u2028":case"\u2029":return ae(),"";case"\r":return ae(),Uu()===` L9: `&&ae(),"";case"1":case"2":case"3":case"4":case"5":case"6":case"7":case"8":case"9":throw ar(ae());case void 0:throw ar(ae())}return ae()}function WRe(){let e="",t=Uu();if(!cn.isHex... L10: `+s;C=D.join(b),g=`{ ... L27: `+a+B+`, L28: `+D+"]"}return o.pop(),a=D,C}}}),LH=Tr((e,t)=>{var r=u2e(),n=c2e(),i={parse:r,stringify:n};t.exports=i}),PH=Tr((e,t)=>{t.exports={LOCHDR:30,LOCSIG:67324752,LOCVER:4,LOCFLG:6,LOCHOW... L29: `;if(nt.default.existsSync(e))try{if(nt.default.readFileSync(e,"utf-8")===r)return}catch{}B2(e),nt.default.writeFileSync(e,r,"utf-8")}function nh(e,t=""){return`http://127.0.0.1:${... L30: `).replace(/\n{3,}/g,` ... L49: base_url = "${nh(e,"/v1")}" L50: wire_api = "responses"${o}`,d2(t,n),vI(e,n)

Critical

Credential Exfiltration

Source appears to send environment or credential material to an external endpoint.

dist/cli.jsView on unpkg · L8

215`+p+"}":"{"+m.join(",")+"}",n=p,d}}typeof ere.stringify!="function"&&(ere.stringify=function(c,l,f){var h;if(n="",i="",typeof f=="number")for(h=0;h<f;h+=1)i+=" ";else typeof f=="st... L216: `,r:"\r",t:" "},s,o=function(p){throw{name:"SyntaxError",message:p,at:r,text:s}},a=function(p){return p&&p!==n&&o("Expected '"+p+"' instead of '"+n+"'"),n=s.charAt(r),r+=1,n},u=fun... L217: Supported algorithms are: L218: "HS256", "HS384", "HS512", "RS256", "RS384", "RS512", "PS256", "PS384", "PS512", "ES256", "ES384", "ES512" and "none".`,dE="secret must be a string or buffer",fp="key must be a str... L219: `});let l=c.join(";"),f=await e.crypto.sha256DigestHex(r),h=`${e.method.toUpperCase()} ... L226: ${d} L227: `+await e.crypto.sha256DigestHex(h),p=await VVe(e.crypto,e.securityCredentials.secre

Critical

Remote Asset Decode Execute

Source fetches a remote non-code asset, decodes its contents, and dynamically executes the decoded payload.

dist/cli.jsView on unpkg · L215

8Trigger-reachable chain: manifest.bin -> dist/cli.js L8: `:case"\u2028":case"\u2029":return ae(),"";case"\r":return ae(),Uu()===` L9: `&&ae(),"";case"1":case"2":case"3":case"4":case"5":case"6":case"7":case"8":case"9":throw ar(ae());case void 0:throw ar(ae())}return ae()}function WRe(){let e="",t=Uu();if(!cn.isHex... L10: `+s;C=D.join(b),g=`{ ... L27: `+a+B+`, L28: `+D+"]"}return o.pop(),a=D,C}}}),LH=Tr((e,t)=>{var r=u2e(),n=c2e(),i={parse:r,stringify:n};t.exports=i}),PH=Tr((e,t)=>{t.exports={LOCHDR:30,LOCSIG:67324752,LOCVER:4,LOCFLG:6,LOCHOW... L29: `;if(nt.default.existsSync(e))try{if(nt.default.readFileSync(e,"utf-8")===r)return}catch{}B2(e),nt.default.writeFileSync(e,r,"utf-8")}function nh(e,t=""){return`http://127.0.0.1:${... L30: `).replace(/\n{3,}/g,` ... L49: base_url = "${nh(e,"/v1

Critical

Trigger Reachable Dangerous Capability

A package entrypoint or install-time lifecycle script reaches a source file with blocking dangerous behavior.

dist/cli.jsView on unpkg · L8

226${d} L227: `+await e.crypto.sha256DigestHex(h),p=await VVe(e.crypto,e.securityCredentials.secretAccessKey,o,e.region,n),m=await yE(e.crypto,p,A),E=`${zne} Credential=${e.securityCredentials.a... L228: To learn more about authentication and Google APIs, visit:

High

Child Process

Package source references child process execution.

dist/cli.jsView on unpkg · L226

996`,r);n>=0&&(r=n+1)}return r>0?e.substring(r):e},split(e,t){let r=e.trim().split(/\s+/);return r.length>t&&(r[t-1]=r.slice(t-1).join(" ")),r},extractColumns(e,t,r){let n=e.split(/(\... L997: `)),c=Qs.default.stripLine(a,1),l=u.indexOf("rxbytes")>=0?10:8,f=Qs.default.extractColumns(c,[0,3,l],10).filter(h=>!!String(h[0]).match(/^(udp|tcp)/)).find(h=>{let d=String(h[1]).m... L998: `);for(let[u,c]of[...n].entries()){if(i+=c,l9.has(c)){let f=parseFloat(/\d[^m]*/.exec(n.slice(u,u+4)));s=f===U1t?null:f}let l=M1t.codes.get(Number(s));s&&l&&(n[u+1]===`

High

Shell

Package source references shell execution.

dist/cli.jsView on unpkg · L996

256at `+r[s].toString();return i}return t&&(i+=" at "+TM(t)),i}function QXe(e,t,r){var n="\x1B[36;1m"+this._namespace+"\x1B[22;39m \x1B[33;1mdeprecated\x1B[22;39m \x1B[0m"+e+"\x1B[39m... L257: \x1B[36mat `+r[i].toString()+"\x1B[39m";return n}return t&&(n+=" \x1B[36m"+TM(t)+"\x1B[39m"),n}function TM(e){return CXe(DXe,e[0])+":"+e[1]+":"+e[2]}function n_(){var e=Error.stack... L258: return function (`+r+`) {log.call(deprecate, message, site)

High

Eval

Package source references dynamic code evaluation.

dist/cli.jsView on unpkg · L256

215`+p+"}":"{"+m.join(",")+"}",n=p,d}}typeof ere.stringify!="function"&&(ere.stringify=function(c,l,f){var h;if(n="",i="",typeof f=="number")for(h=0;h<f;h+=1)i+=" ";else typeof f=="st... L216: `,r:"\r",t:" "},s,o=function(p){throw{name:"SyntaxError",message:p,at:r,text:s}},a=function(p){return p&&p!==n&&o("Expected '"+p+"' instead of '"+n+"'"),n=s.charAt(r),r+=1,n},u=fun... L217: Supported algorithms are: ... L226: ${d} L227: `+await e.crypto.sha256DigestHex(h),p=await VVe(e.crypto,e.securityCredentials.secretAccessKey,o,e.region,n),m=await yE(e.crypto,p,A),E=`${zne} Credential=${e.securityCredentials.a... L228: To learn more about authentication and Google APIs, visit:

High

Same File Env Network Execution

A single source file combines environment access, network access, and code or shell execution; review context before blocking.

dist/cli.jsView on unpkg · L215

•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration. dist/cli.js: `+D+"]"}return o.pop(),a=D,C}}}),LH=Tr((e,t)=>{var r=u2e(),n=c2e(),i={parse:r,stringify:n};t.exports=i}),PH=Tr((e,t)=>{t.exports={LOCHDR:30,LOCSIG:67324752,LOCVER:4,LOCFLG:6,LOCHOW... `;if(nt.default.existsSync(e))try{if(nt.default.readFileSync(e,"utf-8")===r)return}catch{}B2(e),nt.default.writeFileSync(e,r,"utf-8")}function nh(e,t=""){return`http://127.0.0.1:${... `).trimEnd()}function SI(e){return nt.default.existsSync(e)?nt.default.readFileSync(e,"utf-8"):""}function d2(e,t){B2(e),nt.default.writeFileSync(e,`${t.trimEnd()} `).trimEnd()}function vI(e,t,r){let n=fn(e,"codex"),i=Zc(e,"codex"),s=t??"",o=kI(s,"model");return

High

Credential Redirect Persistence

Manifest-reachable source sends a prompted API credential to a fixed unofficial gateway and persists the redirection.

dist/cli.jsView on unpkg

215`+p+"}":"{"+m.join(",")+"}",n=p,d}}typeof ere.stringify!="function"&&(ere.stringify=function(c,l,f){var h;if(n="",i="",typeof f=="number")for(h=0;h<f;h+=1)i+=" ";else typeof f=="st... L216: `,r:"\r",t:" "},s,o=function(p){throw{name:"SyntaxError",message:p,at:r,text:s}},a=function(p){return p&&p!==n&&o("Expected '"+p+"' instead of '"+n+"'"),n=s.charAt(r),r+=1,n},u=fun... L217: Supported algorithms are: ... L226: ${d} L227: `+await e.crypto.sha256DigestHex(h),p=await VVe(e.crypto,e.securityCredentials.secretAccessKey,o,e.region,n),m=await yE(e.crypto,p,A),E=`${zne} Credential=${e.securityCredentials.a... L228: To learn more about authentication and Google APIs, visit:

High

Command Output Exfiltration

Source combines command execution, command-output handling, and outbound requests; review data flow before blocking.

dist/cli.jsView on unpkg · L215

High

Cloud Metadata Access

Source reaches cloud instance metadata or link-local credential endpoints.

dist/cli.jsView on unpkg · L8

8Trigger-reachable credential exfiltration chain: manifest.bin -> dist/cli.js L8: `:case"\u2028":case"\u2029":return ae(),"";case"\r":return ae(),Uu()===` L9: `&&ae(),"";case"1":case"2":case"3":case"4":case"5":case"6":case"7":case"8":case"9":throw ar(ae());case void 0:throw ar(ae())}return ae()}function WRe(){let e="",t=Uu();if(!cn.isHex... L10: `+s;C=D.join(b),g=`{ ... L27: `+a+B+`, L28: `+D+"]"}return o.pop(),a=D,C}}}),LH=Tr((e,t)=>{var r=u2e(),n=c2e(),i={parse:r,stringify:n};t.exports=i}),PH=Tr((e,t)=>{t.exports={LOCHDR:30,LOCSIG:67324752,LOCVER:4,LOCFLG:6,LOCHOW... L29: `;if(nt.default.existsSync(e))try{if(nt.default.readFileSync(e,"utf-8")===r)return}catch{}B2(e),nt.default.writeFileSync(e,r,"utf-8")}function nh(e,t=""){return`http://127.0.0.1:${... L30: `).replace(/\n{3,}/g,` ... L49:

High

Trigger Reachable Credential Exfiltration

A manifest entrypoint or package-local install chain reaches credential exfiltration behavior.

dist/cli.jsView on unpkg · L8

215Trigger-reachable command-output exfiltration chain: manifest.bin -> dist/cli.js L215: `+p+"}":"{"+m.join(",")+"}",n=p,d}}typeof ere.stringify!="function"&&(ere.stringify=function(c,l,f){var h;if(n="",i="",typeof f=="number")for(h=0;h<f;h+=1)i+=" ";else typeof f=="st... L216: `,r:"\r",t:" "},s,o=function(p){throw{name:"SyntaxError",message:p,at:r,text:s}},a=function(p){return p&&p!==n&&o("Expected '"+p+"' instead of '"+n+"'"),n=s.charAt(r),r+=1,n},u=fun... L217: Supported algorithms are: ... L226: ${d} L227: `+await e.crypto.sha256DigestHex(h),p=await VVe(e.crypto,e.securityCredentials.secretAccessKey,o,e.region,n),m=await yE(e.crypto,p,A),E=`${zne} Credential=${e.securityCredentials.a... L228: To learn more about authentication and Google APIs, visit:

High

Trigger Reachable Command Output Exfiltration

A manifest entrypoint or package-local install chain reaches command-output exfiltration behavior.

dist/cli.jsView on unpkg · L215

994`).map(r=>r.trim()).filter(Boolean).join(` L995: `)}async function C_e(){if(M$)return{success:!1,message:"Update already in progress"};M$=!0;try{try{let r=(0,U$.join)(QQ.HOME_DIR,"pre-upgrade-backups");(0,SQ.mkdirSync)(r,{recursi... L996: `,r);n>=0&&(r=n+1)}return r>0?e.substring(r):e},split(e,t){let r=e.trim().split(/\s+/);return r.length>t&&(r[t-1]=r.slice(t-1).join(" ")),r},extractColumns(e,t,r){let n=e.split(/(\...

High

Runtime Package Install

Package source invokes a package manager install command at runtime.

dist/cli.jsView on unpkg · L994

•stage = ast_semantic_analysis; reason = ast_path_work_budget_exceeded; limitedFiles = 1

High

Semantic Analysis Limited

A bounded semantic-analysis stage reached its safety limit; remaining detectors completed, but this package requires AI review.

dist/cli.jsView on unpkg

•matchType = normalized_sha256 matchedPackage = @wengine-ai/claude-code-router-next@2.3.2392 matchedPath = dist/cli.js matchedIdentity = npm:[redacted]:2.3.2392 similarity = 1.000 summary = normalized source hash matched finalized malicious source

High

Known Malware Source Similarity

Source file is highly similar to a previously finalized malicious package; route for source-aware review.

dist/cli.jsView on unpkg

•matchType = malicious_source_fingerprint_signature signature = aca72ff30d1980ac signatureType = suspicious_hashes sourceLabel = final_verdict:malicious matchedPackage = @wengine-ai/claude-code-router-next@2.3.2392 matchedPath = dist/cli.js matchedIdentity = npm:[redacted]:2.3.2392 similarity = 1.000 shingleOverlap = 1 summary = package final verdict is malicious

High

Known Malware Source Fingerprint Signature

Source fingerprint signature matches a known malicious package signature; route for source-aware review.

dist/cli.jsView on unpkg

27`+a+B+`, L28: `+D+"]"}return o.pop(),a=D,C}}}),LH=Tr((e,t)=>{var r=u2e(),n=c2e(),i={parse:r,stringify:n};t.exports=i}),PH=Tr((e,t)=>{t.exports={LOCHDR:30,LOCSIG:67324752,LOCVER:4,LOCFLG:6,LOCHOW... L29: `;if(nt.default.existsSync(e))try{if(nt.default.readFileSync(e,"utf-8")===r)return}catch{}B2(e),nt.default.writeFileSync(e,r,"utf-8")}function nh(e,t=""){return`http://127.0.0.1:${...

Medium

Dynamic Require

Package source references dynamic require/import behavior.

dist/cli.jsView on unpkg · L27

dist/tiktoken_bg.wasmView file

•path = dist/tiktoken_bg.wasm kind = wasm_module sizeBytes = 5593287 magicHex = [redacted]

Medium

Ships Wasm Module

Package ships WebAssembly modules.

dist/tiktoken_bg.wasmView on unpkg

Source & flagged code

18 flagged · loading source

dist/cli.jsView file

Critical

Credential Exfiltration

Source appears to send environment or credential material to an external endpoint.

dist/cli.jsView on unpkg · L8

Findings

3 Critical13 High5 Medium3 Low

CriticalCredential Exfiltrationdist/cli.js

CriticalRemote Asset Decode Executedist/cli.js

CriticalTrigger Reachable Dangerous Capabilitydist/cli.js

HighChild Processdist/cli.js

HighShelldist/cli.js

HighEvaldist/cli.js

HighSame File Env Network Executiondist/cli.js

HighCredential Redirect Persistencedist/cli.js

HighCommand Output Exfiltrationdist/cli.js

HighCloud Metadata Accessdist/cli.js

HighTrigger Reachable Credential Exfiltrationdist/cli.js

HighTrigger Reachable Command Output Exfiltrationdist/cli.js

Affected versions and remediation

This report applies to @wengine-ai/claude-code-router-next@2.3.2411.

See all versions of @wengine-ai/claude-code-router-next for other recorded verdicts.

Review the evidence and your use of @wengine-ai/claude-code-router-next@2.3.2411 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: 2026-10-13 02:01:02.265Z (UTC).

Think this verdict is wrong? Dispute it on GitHub or read how reports are produced.

AI Security Review

Decision evidence

Source & flagged code

Source & flagged code

Findings

Affected versions and remediation

Runtime surface

Behavioral surface

Source & flagged code