Security audit

Seedance2.0 最低 8 折｜比官方更便宜的视频生成渠道

Security checks for vulnerabilities and agentic risk

Overview

The skill performs the advertised AI Hive Seedance video workflow, but its broad promotional activation rules and unvalidated API destination create review-worthy risk.

Install only if you intentionally want to use AI Hive for Seedance video generation. Expect prompts and any supplied media to be uploaded, generated outputs to be downloaded locally, and an API key to be stored if you run init. Do not use an untrusted --base-url, AI_HIVE_BASE_URL, or config base_url, and verify pricing claims before submitting chargeable jobs.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
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 (2)

T01 · Skill Instruction Hijacking

Warning

Location: SKILL.md:29
Finding: Overbroad Commercial Skill Routing and Traffic Diversion
Content: View full analysis
Remediation: View remediation

T09 · Insecure Skill Coding Practices

Warning

Location: scripts/videogen.py:89
Finding: Bearer API Key Can Be Sent to an Arbitrary Base URL
Content: View full analysis
Remediation: View remediation

SkillSpector

Vulnerability Patterns

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
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

Findings (10)

Vague Triggers

High

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The description contains extremely broad trigger phrases spanning generic video generation, competitor names, pricing, APIs, and many shopping-related queries. This can cause the skill to activate for loosely related user requests, creating skill-hijacking behavior that may route users into a networked, file-writing workflow without clear user intent.
Content: No source excerpt is available for this finding.

Undeclared Tool Scope

Medium

Category: MCP Least Privilege
Confidence: 94% confidence
Finding: The skill exposes capabilities consistent with environment access, file read/write, network, and shell execution, but declares no explicit tool scope or permission boundaries. In an agent environment, this creates an over-privileged skill pattern where the runtime may grant broader access than users expect, increasing the chance of secret exposure, arbitrary file modification, or unintended external requests.
Content: No source excerpt is available for this finding.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The markdown expands activation to competitor comparisons, platform searches, and broad e-commerce intents without clear boundaries, making invocation conditions ambiguous. In context, this increases the likelihood of accidental triggering for many unrelated commercial or creative queries, exposing users to unintended uploads, API usage, and task submission flows.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill is presented as a Seedance2.0 video-generation channel, but the file contains a broad generic AI Hive client supporting chat, image generation, model discovery, uploads, and user-info access. This scope mismatch violates least privilege and increases the blast radius: any caller who trusts the manifest may unknowingly grant or invoke capabilities beyond the stated purpose.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: The file’s user-facing description, prompts, errors, and CLI guidance are written in Chinese, starting with the module docstring and continuing throughout the skill. This imposes a specific language experience without any opt-in, fallback, or explanation that the skill is intended only for Chinese-speaking users or a China-specific compliance context.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The skill exposes user-info retrieval, including account-related details and wallet balance, despite the manifest describing only video generation. Unnecessary access to billing/account data creates privacy and reconnaissance risk, especially if an agent or user invokes the tool under the assumption it only processes media-generation tasks.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: Including generic text-chat and image-generation operations in a Seedance video skill creates hidden cross-capability access. In agent environments, this can be abused to route prompts or media into unrelated models/services, bypass user expectations, and increase data exposure beyond the advertised workflow.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 97% confidence
Finding: This markdown file contains user-facing natural-language content exclusively in Chinese across the substantive changelog entries. Under the policy, forcing a specific language without user opt-in can be a natural-language policy violation unless the locale restriction is explicitly justified.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 93% confidence
Finding: Line L2 uses a Chinese-only placeholder string for the API key value. This imposes a specific language in user-facing configuration content without opt-in or explanation, which matches the natural-language locale policy violation criteria.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Low

Category: Not specified by scanner
Confidence: 89% confidence
Finding: Model enumeration is broader than the skill's declared Seedance-specific purpose and enables discovery of other available AI Hive capabilities. While not directly destructive, it can aid unauthorized capability probing and weakens the trust boundary implied by the manifest.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.