registry / AI-agent capability abuse / anyray-connect / 0.11.201

anyray-connect@0.11.201

Anyray connect — points local coding tools (Claude Code, Cursor, Windsurf, SDKs) at the Anyray gateway through supported configuration and enrolled personal authentication.

AI Security Review

scanned 2mo ago · by lpm-firewall-ai

The source establishes a real user-invoked credential-routing and AI-agent configuration capability, but not automatic install-time malware behavior. The package has no registry install lifecycle hook or self-dependency chain.

Static reason

High-risk behavior combination matched malicious policy.

Trigger

A user runs the anyray-connect command.

Impact

The configured tools may route requests and personal gateway credentials through the chosen gateway.

Mechanism

AI-tool configuration rewriting and gateway API-key requests.

AI rationale

Evidence

dist/index.js~/.anyray/connect.json

Network endpoints1

${gateway}/connect/policy

Decision evidence

public snapshot

AI called this Suspicious at 88.0% confidence as Dangerous Capability with low false-positive risk.

Evidence for warning

The executable CLI is the package entrypoint and performs user-invoked tool configuration.
The bundled code stores state under the user's .anyray directory.
It sends an API key header to the selected gateway policy endpoint.

Evidence against

package.json has no install, preinstall, or postinstall hook.
There is no runtime dependency on this package itself.
The observed behavior is activated by running the CLI, not by installation.

Behavioral surface

Source

ChildProcessEnvironmentVarsFilesystemNetworkShell

Supply chain

HighEntropyStringsMinifiedPossibleObfuscationUrlStrings

ManifestNo manifest risk signals triggered.

scanned 1 file(s), 841 KB of source, external domains: api.github.com, api.githubcopilot.com, app.anyray.ai, github.com, gw.acme.com, opencode.ai

Source & flagged code

10 flagged

dist/index.jsView file

1Source sends the broad process environment to a literal external destination. L1: #!/usr/bin/env node L2: var DT="http://localhost:8787",js="anyray-placeholder",bb=/^ark_[A-Za-z0-9_-]{20,512}$/,LT="ark_svc_",NT=/^ark_svc_[A-Za-z0-9_-]{20,512}$/,Pe=e=>{let t=e?.trim();return!!(t&&bb.tes... L3: `;if(a||n.tombstoneAbsent!==!1)try{await Sb(r,s)}catch(i){if(i.code!=="EEXIST")throw i}}let o=`${e}.anyray-tmp-${process.pid}-${Ob()}`;try{await Sb(o,`${JSON.stringify(t,null,2)} ... L5: `,{mode:384})}catch(c){throw await Sa(e,{recursive:!0,force:!0}).catch(()=>{}),c}return async()=>{try{JSON.parse(await Js(o,"utf-8")).nonce===s&&await Sa(e,{recursive:!0,force:!0})... L6: `,{mode:384}),await cf(r,384);try{await Ae(r,n)}catch(o){throw await Sa(r,{force:!0}).catch(()=>{}),o}},de=async e=>{if(ol.getStore

Critical

Hardcoded Runtime Data Exfiltration

Source sends credentials or rich application records to a package-controlled external receiver enabled by default.

dist/index.jsView on unpkg · L1

High

Child Process

Package source references child process execution.

dist/index.jsView on unpkg · L1

115`).map(gB).join(" + [Environment]::NewLine + "),bB=(e={})=>{let t={...yB,...e},n=()=>t.platform()==="win32",r=async()=>{let i=(await fi(ft(),Fn))?.find(c=>c.startsWith("export COPI... L116: `)):K&&ee&&await Ni(O,{force:!0}),await Zl(c,V,Fn)}catch(we){try{X!==void 0&&ee?await ke(O,X):ee&&await Ni(O,{force:!0})}catch{throw new Error("Copilot CLI environment update faile... L117: import { spawn } from "node:child_process"

High

Shell

Package source references shell execution.

dist/index.jsView on unpkg · L115

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/index.jsView on unpkg · L1

•Manifest-reachable source captures an API credential, sends it to a fixed unofficial gateway, and persists that redirection in agent or shell configuration. dist/index.js: var DT="http://localhost:8787",js="anyray-placeholder",bb=/^ark_[A-Za-z0-9_-]{20,512}$/,LT="ark_svc_",NT=/^ark_svc_[A-Za-z0-9_-]{20,512}$/,Pe=e=>{let t=e?.trim();return!!(t&&bb.tes... `),await Ae(o,e),await VT(e,384),await JT(e)}finally{await cn(o,{force:!0}).catch(()=>{})}},Tb=1,Ib=3e4,XT=5e3,zT=/^[a-f0-9]{64}$/,Pb=e=>`${e}.pending`,qT=e=>`${e}.lock`,zp=e=>{if(... `)},In=(e,t={})=>{let n=e.byoUpstream,r=!n&&ae(e,"anthropic"),o=be({...e.metadata,tool:e.metadata.tool??"claude-code"}),a=t.includeClientKey!==!1,s=t.includeAuthToken!==!1,{context... `:"";return{text:a,outcome:a===e?"unchanged":"changed"}};import{readFile as

High

Credential Redirect Persistence

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

dist/index.jsView on unpkg

51WantedBy=timers.target L52: `,RD=(e=Dl())=>[e.command,...e.args].map(t=>/\s/.test(t)?`"${t}"`:t).join(" "),oi=e=>["Auto-refresh scheduling is not wired for this OS. To keep your gateway","key from expiring mi... L53: ${t.stderr??""}`}},MD=async e=>{try{return(await PD(e)).isDirectory()}catch{return!1}},BD=e=>{if(/^Signature=adhoc$/m.test(e))return;let n=/^TeamIdentifier=(.+)$/m.exec(e)?.[1]?.tr... L54: `),o=Po(r),a=1;for(;a<r.length;){let s=o[a];if(s.kind==="table"||s.kind==="key"&&!ZD(r[a],s.key))break;a+=1}for(;a>1&&r[a-1].trim()==="";)a-=1;return`${n.replace(/\n+$/,`

High

Command Output Exfiltration

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

dist/index.jsView on unpkg · L51

1#!/usr/bin/env node L2: var DT="http://localhost:8787",js="anyray-placeholder",bb=/^ark_[A-Za-z0-9_-]{20,512}$/,LT="ark_svc_",NT=/^ark_svc_[A-Za-z0-9_-]{20,512}$/,Pe=e=>{let t=e?.trim();return!!(t&&bb.tes... L3: `;if(a||n.tombstoneAbsent!==!1)try{await Sb(r,s)}catch(i){if(i.code!=="EEXIST")throw i}}let o=`${e}.anyray-tmp-${process.pid}-${Ob()}`;try{await Sb(o,`${JSON.stringify(t,null,2)} ... L5: `,{mode:384})}catch(c){throw await Sa(e,{recursive:!0,force:!0}).catch(()=>{}),c}return async()=>{try{JSON.parse(await Js(o,"utf-8")).nonce===s&&await Sa(e,{recursive:!0,force:!0})... L6: `,{mode:384}),await cf(r,384);try{await Ae(r,n)}catch(o){throw await Sa(r,{force:!0}).catch(()=>{}),o}},de=async e=>{if(ol.getStore()===!0){await af(e);return}let t=await uf();try{... L7: `)},In=(e,t={})=>{let

High

Cloud Metadata Access

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

dist/index.jsView on unpkg · L1

51Trigger-reachable command-output exfiltration chain: scripts.start -> dist/index.js L51: WantedBy=timers.target L52: `,RD=(e=Dl())=>[e.command,...e.args].map(t=>/\s/.test(t)?`"${t}"`:t).join(" "),oi=e=>["Auto-refresh scheduling is not wired for this OS. To keep your gateway","key from expiring mi... L53: ${t.stderr??""}`}},MD=async e=>{try{return(await PD(e)).isDirectory()}catch{return!1}},BD=e=>{if(/^Signature=adhoc$/m.test(e))return;let n=/^TeamIdentifier=(.+)$/m.exec(e)?.[1]?.tr... L54: `),o=Po(r),a=1;for(;a<r.length;){let s=o[a];if(s.kind==="table"||s.kind==="key"&&!ZD(r[a],s.key))break;a+=1}for(;a>1&&r[a-1].trim()==="";)a-=1;return`${n.replace(/\n+$/,`

High

Trigger Reachable Command Output Exfiltration

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

dist/index.jsView on unpkg · L51

High

Base64 Obscured Url

Source decodes a Base64-obscured HTTP endpoint at runtime.

dist/index.jsView on unpkg · L1

•stage = ast_semantic_analysis; reason = ast_alias_growth_limit_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/index.jsView on unpkg

Source & flagged code

10 flagged · loading source

dist/index.jsView file

Critical

Hardcoded Runtime Data Exfiltration

Source sends credentials or rich application records to a package-controlled external receiver enabled by default.

dist/index.jsView on unpkg · L1

High

Child Process

Package source references child process execution.

dist/index.jsView on unpkg · L1

Findings

1 Critical9 High3 Medium5 Low

CriticalHardcoded Runtime Data Exfiltrationdist/index.js

HighChild Processdist/index.js

HighShelldist/index.js

HighSame File Env Network Executiondist/index.js

HighCredential Redirect Persistencedist/index.js

HighCommand Output Exfiltrationdist/index.js

HighCloud Metadata Accessdist/index.js

HighTrigger Reachable Command Output Exfiltrationdist/index.js

HighBase64 Obscured Urldist/index.js

HighSemantic Analysis Limiteddist/index.js

MediumNetwork

MediumEnvironment Vars

MediumStructural Risk Force Deep Review

Affected versions and remediation

This report applies to anyray-connect@0.11.201.

See all versions of anyray-connect for other recorded verdicts.

Review the evidence and your use of anyray-connect@0.11.201 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: 2026-09-05 07:09:49.141Z (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