Back to skill

Security audit

seedance-anime-drama

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly coherent for generating anime videos through Ofox, but it includes an under-disclosed path that may require making user images publicly reachable and a mutable npx recovery install path.

Install only if you are comfortable sending prompts, generated images, and video requests to Ofox and its upstream providers with an OFOX_API_KEY. Do not use the identity-reference path with personal or sensitive images unless you can provide a private, temporary URL and delete it afterward; prefer the local `--frame-first-image` path where possible. Treat the npx recovery command as a supply-chain step and pin or verify the source before using it.

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
SKILL.md:892
Finding
Unpinned Third-Party Installation Through npx## Vulnerability Details **File Location**: `SKILL.md`, line 892 **Vulnerability Type**: Unpinned and unaudited third-party dependency execution **Risk Level**: Medium ### Vulnerable Code Snippet ```text Fix: `npx skills add ofoxai/skills` (the whole repo). ``` ### Technical Analysis The recovery instructions recommend invoking `npx` to install an entire external repository. The command does not pin an immutable package version or source commit and does not require checksum or signature verification. Because the external package and the referenced image and video core scripts are absent from this artifact, their lifecycle behavior, network destinations, credential handling, filesystem operations, and transitive dependencies cannot be audited here. The instruction also installs the whole repository rather than only the two core components required by this Skill, unnecessarily increasing the supply-chain attack surface. This is an insecure dependency pattern rather than evidence that the current upstream is malicious. Exploitation would require the package, its distribution channel, or one of its dependencies to become compromised or resolve to attacker-controlled content. ### Attack Path 1. The expected sibling core scripts are missing. 2. An Agent follows the documented recovery instruction. 3. `npx` resolves mutable external package content without an immutable version or integrity constraint. 4. A compromised package, registry account, distribution channel, or transitive dependency supplies malicious code. 5. The retrieved code executes during installation or subsequent Skill use with the Agent process's privileges. 6. That code may access data available to the process, including `OFOX_API_KEY`, user-provided media, generated artifacts, and writable project files. ### Impact Assessment Successful exploitation could provide arbitrary code execution with the privileges of the user running the Agent. The resulting scope may include: - Reading and exfi ...[truncated 419 chars]
Remediation
## Remediation Suggestions 1. Pin the dependency to an audited, immutable package version and source commit rather than resolving the latest mutable content. 2. Publish and verify cryptographic checksums or signatures before installation. 3. Use an explicit trusted registry and disable unexpected lifecycle scripts where supported. 4. Install only the required `ofox-image-core` and `ofox-video-core` components instead of the entire repository. 5. Commit a lockfile covering direct and transitive dependencies. 6. Vendor the required audited scripts into the distributed package when licensing and maintenance constraints permit. 7. Document the expected installed files and verify their hashes before execution. 8. Run installation and generation components with least privilege and a restricted environment that exposes only `OFOX_API_KEY` and the files necessary for the current task.
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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
Findings (1)

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill instructs the operator to host a user's local image at a publicly reachable HTTPS URL in order to use `input_references`, but it does so without any explicit privacy warning, consent requirement, retention guidance, or recommendation for temporary/private hosting. This can expose sensitive user images to unintended third parties, indexing, or persistent public access, especially because the content may be personal character art or user-provided photos and the operator is guided to make it externally reachable.

Static analysis

No suspicious patterns detected.