Back to skill

Security audit

seedance-short-drama

Security checks for vulnerabilities and agentic risk

Overview

This video-generation skill is mostly transparent, but it depends on installing an unpinned external core package that will later use an Ofox API key for paid jobs.

Review the external `ofox-video-core` code and install it from a trusted pinned revision before exposing `OFOX_API_KEY`. Use the documented dry-run and approval gate, set an explicit output directory, and do not let the agent submit paid jobs until the quoted prompt and cost match what you want.

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:778
Finding
Unpinned Remote Package Execution and Skill Installation## Vulnerability Details **File Location**: `SKILL.md`, lines 778–790 **Vulnerability Type**: Supply-chain exposure through unpinned package execution and mutable remote content **Risk Level**: Medium ### Vulnerable Code ```text bash: ../ofox-video-core/references/ofox-video.sh: No such file or directory ``` ```text This means `ofox-video-core` isn't installed alongside this skill — not that anything is broken. This skill delegates all execution to it and reaches it by relative path. Fix: `npx skills add ofoxai/skills` (the whole repo). ``` ### Technical Analysis The documented recovery procedure tells the Agent to invoke `npx skills add ofoxai/skills`. It does not pin the `skills` npm package to an exact version or pin the `ofoxai/skills` repository content to an audited release or commit. No integrity digest or signature-verification step is specified. Consequently, the code executed by `npx` and the Skill content installed from the remote source may differ from the content that existed when this project was audited. The installed `ofox-video-core` shell script is subsequently intended to operate with access to `OFOX_API_KEY` and to submit, poll, and download paid video-generation jobs. This is a supply-chain weakness rather than evidence that the current package or repository is malicious. The audited project contains only documentation; the delegated core implementation is absent and could not be reviewed. ### Attack Path 1. The local installation lacks the sibling `ofox-video-core` Skill. 2. The Agent encounters the documented missing-file error. 3. Following `SKILL.md`, the Agent runs `npx skills add ofoxai/skills`. 4. `npx` resolves and executes an unpinned version of the installer package. 5. The installer retrieves mutable remote Skill content without requiring a verified commit or integrity digest. 6. If either distribution channel is compromised, attacker-controlled code can be installed as `ofox-video-core`. 7. The Agent later invokes the i ...[truncated 862 chars]
Remediation
## Remediation Suggestions 1. Pin the `skills` npm package to a reviewed exact version rather than relying on the version resolved by default through `npx`. 2. Pin `ofoxai/skills` to an audited immutable commit hash or signed release. 3. Publish and verify cryptographic checksums or signatures for installed artifacts. 4. Prefer bundling the required `ofox-video-core` implementation with the distributed Skill when licensing and packaging permit. 5. If remote installation remains necessary, download into a staging location and require review or integrity verification before execution. 6. Do not expose `OFOX_API_KEY` to newly installed scripts until their provenance and integrity have been verified. 7. Run installation and delegated scripts with minimum filesystem and network permissions. 8. Document the exact trusted package version, repository revision, and verification procedure in `SKILL.md`.
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

Static analysis

No suspicious patterns detected.