Back to plugin

Security audit

Call My Agent

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent local voice-agent app, but it exposes powerful OpenAI and OpenClaw actions through web endpoints that are not clearly authenticated beyond the private network boundary.

Install only if you will keep this app reachable only by trusted devices and users. Keep the default localhost binding, use Tailscale rather than a public tunnel, set a strong CALL_MY_AGENT_CALL_TOKEN before remote use, and understand that anyone who can reach the app may be able to start voice sessions using your OpenAI access, invoke your OpenClaw agent, schedule reminders, or read/write this app's local capture state.

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep

Endor dependency reachability

Checked Nov 10, 2026 · 10:59 AM UTC

8 reachable vulnerabilities

  • High

    Next.js: Denial of Service in App Router using Server Actions

    next@15.5.18Fixed in 15.5.21

    GHSA-m99w-x7hq-7vfjDirect dependency of @mrbese/call-my-agent@0.1.2
  • Medium

    Next.js builds the cache key for functions marked with the use cache directive in encodeFormData, which turned binary arguments such as Uint8Array, Blob or File values into a UCS-2 string and appended only its length and contents. Distinct binary values, or values with different names and types, could therefore produce identical cache keys, because the key carried no binary marker, file name, type or byte length. An application passing binary arguments to a cached function could return one user's cached result to another, disclosing data or poisoning the cache.

    next@15.5.18Fixed in 16.4.5

    CVE-2026-119349Direct dependency of @mrbese/call-my-agent@0.1.2
  • Critical

    Next.js: Unauthenticated Remote Code Execution in Image Optimization API when AVIF files are used

    next@15.5.18Fixed in 15.5.24

    GHSA-2xp9-vwfh-vxw4Direct dependency of @mrbese/call-my-agent@0.1.2
  • Critical

    Next.js: Unauthenticated Remote Code Execution on windows-hosted servers

    next@15.5.18Fixed in 15.5.24

    GHSA-p293-qw3h-jr36Direct dependency of @mrbese/call-my-agent@0.1.2
  • High

    Next.js: Server-Side Request Forgery in Server Actions on custom servers

    next@15.5.18Fixed in 15.5.21

    GHSA-89xv-2m56-2m9xDirect dependency of @mrbese/call-my-agent@0.1.2
  • Critical

    Next.js: Server-Side Request Forgery in rewrites via attacker-controlled destination hostname

    next@15.5.18Fixed in 15.5.21

    GHSA-p9j2-gv94-2wf4Direct dependency of @mrbese/call-my-agent@0.1.2
  • Medium

    HttpOnly cookie exposure through internal response header on Next.js Edge runtime routes

    next@15.5.18Fixed in 15.5.29

    CVE-2026-119346Direct dependency of @mrbese/call-my-agent@0.1.2
  • Medium

    Self-hosted Next.js servers copy the headers of a fetched upstream response onto the outgoing response without dropping content-length, content-encoding and transfer-encoding. In NextNodeServer.handleCatchallMiddlewareRequest and sendResponse, a response whose body was transparently decompressed is therefore sent with a stale declared length or encoding that no longer matches the body. Behind an HTTP/1.1 reverse proxy that queues pipelined responses, a crafted upstream response can desynchronise responses between clients, leading to response injection and disclosure of other clients responses.

    next@15.5.18Fixed in 15.5.29

    CVE-2026-119348Direct dependency of @mrbese/call-my-agent@0.1.2