Back to skill

Security audit

Visual QA

Security checks across malware telemetry and agentic risk

Overview

This skill is a coherent visual review workflow with no evidence of hidden behavior, exfiltration, persistence, or unsafe automatic execution.

Reasonable to install if you trust the publisher and installation source. For sensitive projects, prefer a pinned commit or reviewed release, use synthetic screenshots, and avoid giving fix mode authority unless you want the agent to edit code.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
README.md:39
Finding
Unpinned npx Installer and Mutable Skill Source## Vulnerability Details **File Location**: `README.md`, line 39 **Vulnerability Type**: Supply-chain risk caused by unpinned executable dependencies and a mutable source reference **Risk Level**: Medium **Complete Code Snippet**: ```markdown 1. Install the skill: `npx skills add AntreasAntoniou/visual-qa`. ``` ### Technical Analysis The documented installation command invokes the `skills` npm package through `npx` without specifying an exact package version or integrity value. Depending on the local npm environment and cache state, `npx` may download and execute the package version currently selected by the registry. The `AntreasAntoniou/visual-qa` source reference is also not pinned to an immutable commit, signed release, or verified artifact digest. Consequently, the installer implementation and the Skill content installed by this command can differ from the versions reviewed during this audit. This is a supply-chain weakness rather than evidence that the current dependency is malicious. Exploitation requires compromise, malicious publication, or unauthorized modification of one of the mutable upstream components. ### Attack Path 1. An attacker compromises the npm account, package publication process, registry resolution path, or repository used by the installation workflow. 2. The attacker publishes a modified `skills` package or changes the mutable Skill source referenced by `AntreasAntoniou/visual-qa`. 3. A user follows the documented unpinned `npx` command. 4. `npx` resolves and executes the attacker-controlled package version, or the installer retrieves attacker-controlled Skill content. 5. Malicious installer code executes with the privileges of the invoking user, or malicious Skill instructions become active in subsequent agent sessions. ### Impact Assessment Successful compromise of the executable npm dependency could allow arbitrary code execution with the permissions of the user running the installatio ...[truncated 485 chars]
Remediation
## Remediation Suggestions 1. Pin the `skills` CLI to an exact reviewed version, for example by using an exact npm version rather than an unversioned package name. 2. Pin the Skill source to an immutable commit hash or a cryptographically signed release instead of a mutable repository reference. 3. Publish and verify integrity hashes or signatures for released Skill artifacts. 4. Use a lockfile or an equivalent dependency-locking mechanism where the installation workflow supports one. 5. Configure CI to verify package provenance, release signatures, and expected artifact digests. 6. Document that users should inspect the resolved package and Skill content before granting execution or agent tool permissions. 7. Run installation with least privilege in an isolated environment, without production credentials or unnecessary filesystem access.
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

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.dynamic_code_execution

Dynamic code execution detected.

Critical
Code
suspicious.dynamic_code_execution
Location
tests/test_validate_manifest.py:19