Back to skill

Security audit

hub-task-manager

Security checks for vulnerabilities and agentic risk

Overview

ClawScan rates this skill suspicious because a finding reports that it may send a local node credential to a configurable destination, and the scan was only partial.

Consider whether you trust the Hub destination and are comfortable with the skill accessing and transmitting your local node credential. The scan did not fully inspect the skill, so its evidence is incomplete; review the credential use and destination before installing.

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (5)

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: env, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · SKILL.md (reported line 27)May include surrounding context.

1. List Available Tasks

bash
curl -s "https://hub.evomap.ai/a2a/task/list?reputation=0&limit=20" \
  -H "X-Node-Secret: $(cat ~/.evomap/node_secret)"

The skill silently reads a local node credential and sends it to the configured Hub without disclosing this credential use.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

getNodeSecret reads ~/.evomap/node_secret, and the API helpers transmit its contents in the X-Node-Secret header. Although Hub requests are expected for task management, the file provides no warning, explanatory comment, or user-facing notice about accessing this local credential and sending it to the configured destination.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
94% confidence
Finding

Although authenticating to the default Hub is legitimate, A2A_HUB_URL can select an arbitrary destination or plaintext HTTP endpoint, and the code forwards the local node secret without validating either. If an attacker can influence that configuration, normal task operations disclose the credential; interpolating the URL into a shell command also permits shell injection under that condition.

Content

Scanner excerpt · index.js (reported line 21)May include surrounding context.

js
function apiGet(endpoint) {
  const secret = getNodeSecret();
  const cmd = `curl -s "${HUB_URL}${endpoint}" -H "X-Node-Secret: ${secret}"`;
  return JSON.parse(execSync(cmd, { encoding: 'utf-8' }));
}

The manifest promises retry logic, but requests are attempted only once.

Low
Category
Excessive Agency
Confidence
99% confidence
Finding

The manifest explicitly describes task management with retry logic. Both API helpers execute a single curl invocation without retry flags, loops, or recovery handlers, so transient failures are not retried.

Content

No source excerpt is available for this finding.