registry  /  syrax-audit  /  1.0.7

syrax-audit@1.0.7

Herramienta CLI oficial de Syrax para auditar tu código y obtener prompts de reparación de IA.

AI Security Review

scanned 2h ago · by lpm-firewall-ai

Review flagged AI-agent configuration or capability changes. This remains warn-only unless evidence shows foreign-agent hijack through preinstall/install/postinstall, hidden persistence, exfiltration, remote code execution, or other concrete malicious behavior.

Static reason
One or more suspicious static signals were detected.
Trigger
Running `syrax-audit` interactively and accepting Husky or Premium Autofix prompts.
Impact
Can persist a project Git hook and cause an external AI CLI to modify the current project using remotely supplied audit prompts.
Mechanism
explicit Git-hook setup and AI-agent control-file/CLI execution
Rationale
The package exposes real, remotely influenced AI-agent and Git-hook mutation capabilities, but they require explicit interactive user choices and are not hidden lifecycle behavior. This is a warning-level dangerous capability rather than confirmed malicious execution.
Evidence
package.jsonbin/index.js.env.husky/pre-commit.husky/pre-pushsyrax-report.json.cursorrules
Network endpoints2
localhost:5173syrax-analyzer.moisesvalero.es/api/web-audit/cli-analyze

Decision evidence

public snapshot
AI called this Suspicious at 90.0% confidence as Dangerous Capability with low false-positive risk.
Evidence for warning
  • `bin/index.js` POSTs audited local HTML and optional Syrax API key to `syrax-analyzer.moisesvalero.es`.
  • Interactive setup can install Husky and create executable `.husky/pre-commit` or `.husky/pre-push` hooks.
  • Explicit Autofix writes `.cursorrules` directing AI agents to consume remote-supplied issue prompts.
  • Claude Autofix runs `npx @anthropic-ai/claude-code` with instructions to modify the current codebase.
Evidence against
  • `package.json` has no preinstall, install, postinstall, or prepare lifecycle hook.
  • All hook and Autofix mutations are behind interactive user confirmations and agent selection.
  • Source reads only `SYRAX_API_KEY` from `.env`; no general credential harvesting found.
  • No eval, VM, binary loading, obfuscation, or hidden package files found.
Behavioral surface
Source
ChildProcessEnvironmentVarsFilesystemNetworkShell
Supply chain
UrlStrings
ManifestNo manifest risk signals triggered.
scanned 1 file(s), 18.1 KB of source, external domains: syrax-analyzer.moisesvalero.es

Source & flagged code

3 flagged · loading source
bin/index.jsView file
4import path from 'path'; L5: import { execSync } from 'child_process'; L6: import prompts from 'prompts';
High
Child Process

Package source references child process execution.

bin/index.jsView on unpkg · L4
141launchedProcess = spawn(`${packageManager} run dev`, [], { L142: shell: true, L143: stdio: 'pipe'
High
Shell

Package source references shell execution.

bin/index.jsView on unpkg · L141
324const installCmd = L325: packageManager === 'pnpm' ? 'pnpm add -D husky' : 'npm install --save-dev husky'; L326: execSync(installCmd, { stdio: 'ignore' }); L327: execSync('npx husky init', { stdio: 'ignore' });
High
Runtime Package Install

Package source invokes a package manager install command at runtime.

bin/index.jsView on unpkg · L324

Findings

3 High2 Medium3 Low
HighChild Processbin/index.js
HighShellbin/index.js
HighRuntime Package Installbin/index.js
MediumNetwork
MediumEnvironment Vars
LowScripts Present
LowFilesystem
LowUrl Strings