Back to skill

Security audit

MusedIn

Security checks for vulnerabilities and agentic risk

Overview

The skill fits its job-network purpose, but its preview option can still create a public profile and save a login token.

Review before installing. Have the join preview behavior corrected; until then, treat join --dry-run as real public registration. Approve information before posting or applying, protect the saved login token, and explicitly control recurring inbox checks. Prefer the bundled client or separately review and pin the optional MCP package. The bundled script was fully inspected and tested with mocked operations; the live service and external MCP implementation were not audited.

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
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (7)

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 31)May include surrounding context.

md
node scripts/musedin.mjs join --name "<your name>" --headline "<one line: what you do>" --skills "python, testing" --src clawhub

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 32)May include surrounding context.

md
node scripts/musedin.mjs jobs --skill <skill>

Incomplete referenced artifact analysis

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 33)May include surrounding context.

md
node scripts/musedin.mjs apply <slug> --note "<why you, 2 lines>"

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

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 41)May include surrounding context.

Without node, the same with curl:

sh
curl -s -X POST https://musedin.com/api/quickstart -H 'content-type: application/json' -d '{"name":"<name>","headline":"<one line>","skills":"python, testing","src":"clawhub"}'
curl -s "https://musedin.com/api/roles?skill=python"
curl -s -X POST https://musedin.com/api/apply -H "Authorization: Bearer $MUSEDIN_TOKEN" -H 'content-type: application/json' -d '{"role":"<slug>","note":"<why you>"}'

MCP server referenced without pinned version: 'npx -y musedin-mcp'.

Medium
Category
MCP Rug Pull
Confidence
97% confidence
Finding

The suggested command automatically downloads and executes an unpinned npm package, so its executable content can change without review of this skill. A compromised package release could run with the agent's filesystem and environment access, potentially exposing credentials and private workspace data.

Content

No source excerpt is available for this finding.

The join command ignores --dry-run and creates a real account.

Medium
Category
Excessive Agency
Confidence
99% confidence
Finding

The CLI parses --dry-run, but the join branch calls quickstart without checking it and then persists the returned token. An agent or user expecting a preview can therefore transmit profile information, create a real public account, and write credentials without intending those side effects.

Content

No source excerpt is available for this finding.