Back to skill

Security audit

secret-leak-scan-lite

Security checks for vulnerabilities and agentic risk

Overview

The scanners rate this skill suspicious because its installation command is unpinned and the supplied material does not establish that it performs its advertised credential scan; some code was only partially inspected.

Consider not installing until the external installer and skill revision are pinned and reviewed, and the advertised scanning behavior can be verified. The scan covered only half of the components fully; the script and test were partially inspected, so the evidence is incomplete.

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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (5)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · README.md (reported line 7)May include surrounding context.

md
before committing. Masked fingerprints only — raw secret values are never
emitted.

Detects: AWS access key IDs, GitHub personal access tokens, Slack tokens,
PEM private-key blocks.

## Install

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The supplied chunk is an installation command, not the declared static scanner. Repository cloning could support installation of such a scanner, but the repository contents are not supplied, so they cannot establish the advertised behavior. This assessment is limited to the provided code.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied command installs a skill; it does not scan project files or generate a masked JSON report. Installation might be a setup step, but no implementation is supplied to establish the declared scanning behavior or the guarantee that raw secrets are never emitted.

Content

No source excerpt is available for this finding.

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 8)May include surrounding context.

md
horized to analyze** for accidentally committed credentials, using the bundled `scripts/scan.mjs`, and turn the JSON report into a remediation list.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
89% confidence
Finding

Although this is not an MCP server reference, the installation command invokes an unpinned npm package through npx and does not identify an immutable revision of the skill. A compromised or unexpectedly changed upstream release could execute code with the installer's permissions; this is particularly concerning in environments containing credentials.

Content

No source excerpt is available for this finding.