Back to skill

Security audit

keyframe-animation

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed Ofox video-generation workflow that uses user-provided images and an API key with an explicit cost-approval step before paid jobs.

Before installing, confirm you trust the Ofox publisher and are comfortable sending the selected images to Ofox for generation. Prefer a reviewed or pinned installer path for ofox-video-core when possible, keep OFOX_API_KEY scoped to Ofox use, and only approve paid runs after reviewing the dry-run cost table.

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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Rogue AgentSelf-Modification, Session Persistence
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (12)

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 71)May include surrounding context.

md
one this `SKILL.md` sits in. From anywhere else nothing resolves — use the

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 652)May include surrounding context.

md
one this `SKILL.md` sits in. From anywhere else nothing resolves — use the

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
70% confidence
Finding

npx commands without a version suffix (e.g. @1.0.0) create a rug-pull risk if the upstream server is compromised and publishes a malicious update.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
79% confidence
Finding

This line recommends 'npx ofox-skills' as the way to install every skill in the repo so sibling paths resolve. Even in changelog prose, normalizing an unpinned broad installer increases supply-chain risk and can lead a user or agent to fetch and execute whatever package version is current at runtime, with a wider-than-necessary installation scope.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: keyframe-animation
description: Requires OFOX_API_KEY — create one at https://app.ofox.ai, plus two images you already have, a start frame and an end frame. Animates the motion between them as one video — both frames go into a single Ofox video job, the clip opens on A, closes on B, and the model fills the middle. Use when a user has two stills and wants the in-between animated, e.g. "here is the before and the after, animate the transition", "make a video that starts on this image and ends on that one", "tween these two frames", or "move the object from where it sits in the first picture to where it sits in the second". Do not use when only one image exists (animating a single frame is seedance-ad-creative or seedance-product-video), or when the pair is two states of a user interface (see product-demo).
license: MIT
version: "2.0.0"
homepage: https://github.com/ofoxai/skills/tree/main/skills/keyframe-animation

Rp1

Medium
Category
MCP Rug Pull
Confidence
79% confidence
Finding

The skill instructs users to install a dependency via npx skills add ... without pinning a version, which exposes them to supply-chain drift or a compromised latest release. Because this is an installation command that users are explicitly told to run when a prerequisite is missing, the documentation materially increases the chance of executing unreviewed code from the registry.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
83% confidence
Finding

This line recommends npx ofox-skills ofox-video-core without a fixed version, so anyone following the instructions will execute whichever package version is current at the time. If the package or one of its transitive dependencies is compromised, that can lead to arbitrary code execution on the user's machine during installation or startup.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The changelog states that installing every skill in the repo is what keeps sibling paths resolvable, which effectively endorses an overbroad installation path without presenting a narrower default. That can push users or downstream agents toward unnecessary code installation outside the immediate need, increasing attack surface and the blast radius of any compromised package or skill.

Content

No source excerpt is available for this finding.