Back to skill

Security audit

steampunk-transform-video-gen-seedance2-0

Security checks for vulnerabilities and agentic risk

Overview

The skill is mostly disclosed video-generation tooling, but its bundled CLI can upload any readable local file passed as an image to WeryAI without an enforced path, type, or consent boundary.

Install only if you are comfortable giving this skill a WeryAI API key and sending prompts, image URLs, and any explicitly chosen local attachment files to WeryAI. Prefer public HTTPS image URLs. Do not pass local paths unless you have verified the exact file and accept that the script may upload it to a third-party service as a public URL. Run it in a restricted workspace if possible.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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)

T09 · Insecure Skill Coding Practices

Error
Location
scripts/video_gen.js:129
Finding

Unrestricted Local File Read and Upload Through Image Parameters

Content
View full analysis
{ if (typeof value === 'string' && value.trim()) candidates.push(value.trim()); }; const maybePushMany = (value) => { if (Array.isArray(value)) { for (const item of value) maybePush(item); } }; maybePush(data.url); maybePush(data.file_url); maybePush(data.fileUrl); maybePush(data.https_url); maybePush(data.httpsUrl); maybePush(data.public_url); maybePush(data.publicUrl); maybePush(data.object_url); maybePush(data.objectUrl); maybePush(data.origin_url); maybePush(data.originUrl); maybePushMany(data.object_url_list); maybePushMany(data.objectUrlList); if (data.file && typeof data.file === 'object') { candidates.push(...collectPossibleUploadUrls(data.file)); } if (data.data && typeof data.data === 'object') { candidates.push(...collectPossibleUploadUrls(data.data)); } if (Array.isArray(data.files)) { for (const item of data.files) { candidates.push(...collectPossibleUploadUrls(item)); } } if (Array.isArray(data.list)) { for (const item of data.list) { candidates.push(...collectPossibleUploadUrls(item)); } } return candidates; } function extractUploadedFileUrl(res) { const urls = collectPossibleUploadUrls(res ...[truncated 5379 chars]
Remediation
View remediation
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (24)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill is presented as a narrow steampunk Seedance 2.0 transformer, but its documented and implied behavior includes broader video-generation operations, model selection gaps, task management, and local file upload to remote services. That mismatch is dangerous because users and agents may grant trust, secrets, and file access based on the narrow description while the underlying capability surface is substantially wider, enabling unexpected data egress and policy bypass. The surrounding text even admits the script does not enforce the claimed model restriction in code, which strengthens the concern.

Content

No source excerpt is available for this finding.

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 16)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 20)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 21)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 22)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 28)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 29)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 30)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 34)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 64)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

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
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 84)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 98)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 148)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 157)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Ae1

High
Category
analysis-evasion
Confidence
100% confidence
Finding

Referenced artifact was not completely inspected

Content

Scanner excerpt · SKILL.md (reported line 164)May include surrounding context.

md
Node.js 18+. This skill uses **only** `SEEDANCE_2_0`. When you run the CLI, **`scripts/video_gen.js`** must exist; **[`WERYAI_VIDEO_API.md`](resources/WERYAI_V

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill declares runtime capabilities requiring environment secrets and network access, but does not explicitly scope or constrain tool permissions. That makes it easier for an agent or runtime to invoke broader-than-necessary capabilities, including exfiltrating sensitive inputs or using the API key in unintended ways. In this context, the skill also encourages remote uploads and paid API use, so lack of explicit tool scope meaningfully increases risk.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The instructions require that every short or vague user brief be turned into a full English production prompt. This imposes a specific language choice as a mandatory behavior, and the file does not offer the user a language option or opt-in for English-only processing.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The implicit trigger prompt is broad enough that ordinary user requests about steampunk-style video transformation could activate the skill without an explicit invocation boundary. This can cause unintended skill routing, surprising behavior, or policy bypass if the skill is selected in contexts where the user did not clearly request it.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The file states that the script does not enforce skill-specific model allowlists, meaning callers can invoke models outside the manifest’s declared WeryAI Seedance 2.0 scope. This weakens policy boundaries and can lead to unauthorized model use, unexpected capabilities, cost/control bypasses, and behavior inconsistent with the reviewed skill contract.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The documentation explicitly describes a local-file upload path that reads arbitrary local files and sends them to a remote WeryAI upload endpoint before generation. That behavior exceeds the advertised scope of a prompt-to-steampunk-video skill and creates a real risk of unintended local file access and exfiltration if an agent follows the shared docs rather than the skill’s narrower intent.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest says this skill rebuilds briefs into a specific Victorian-industrial steampunk style using WeryAI Seedance 2.0 only. However, the file documents itself as a general-purpose WeryAI video-generation CLI and accepts arbitrary model identifiers and prompts without enforcing any steampunk styling or a Seedance 2.0 model restriction.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
84% confidence
Finding

The file contains natural-language UX instructions for user-facing responses but does not indicate that the user may choose language or locale. Under the stated policy, forcing or implicitly assuming a single language without opt-in can be a natural-language policy concern.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The manifest presents a narrowly scoped skill for steampunk transformation with Seedance 2.0 only. The code includes a models command that fetches the provider's general video model registry, which enables selecting capabilities outside the described single-model, single-style purpose.

Content

No source excerpt is available for this finding.