Back to skill

Security audit

Huo15 Comic Video

Security checks for vulnerabilities and agentic risk

Overview

The skill is a coherent video-generation tool, but it needs Review because unvalidated scene IDs can make it read/upload or write outside the intended folders and its cost controls are weaker than they appear.

Install only if you trust the script JSON and storyboard inputs, understand that frames and prompts are sent to Volcengine Ark using your ARK_API_KEY, and are comfortable reviewing output paths before running. Do not rely on the bundled cost cap as a hard spending guard; set provider-side limits where 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 (2)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/video.py:87
Finding

Path Traversal Through Unvalidated Scene Identifiers

Content
View full analysis
Remediation
View remediation
str: if not isinstance(value, str) or not SID_PATTERN.fullmatch(value): raise ValueError("Scene ID must contain only letters, digits, underscores, or hyphens") return value ``` 2. Resolve and verify every path before reading or writing: ```python def confined_path(base: pathlib.Path, name: str) -> pathlib.Path: base = base.resolve() candidate = (base / name).resolve() if not candidate.is_relative_to(base): raise ValueError(f"Path escapes permitted directory: {candidate}") return candidate ``` 3. Construct paths only after validating the scene identifier: ```python sid = validate_sid(scene["id"]) frame = confined_path(frame_dir, f"{sid}.png") out = confined_path(out_dir, f"{sid}.mp4") last_out = confined_path(last_dir, f"{sid}_last.png") ``` 4. Reject absolute paths, path separators, `.` components, and `..` components even if additional validation is introduced elsewhere. 5. Document the accepted scene-ID format in `SKILL.md` and fail before submitting any network jobs if validation fails. ]]>

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/video.py:101
Finding

Cost Limit Is Enforced Only After Billable API Work Is Submitted

Content
View full analysis
None: """开工前硬熔断. Raises: BudgetExceeded: 估算超 cap 时抛出,附带降级建议字符串。 """ if estimated_total > self.cap: suggestions = self.downgrade_suggestions(estimated_total) raise BudgetExceeded( f"预估成本 ¥{estimated_total:.2f} 超过熔断上限 ¥{self.cap:.2f}。\n" f"降级建议:\n{suggestions}\n" f"如需继续,请提升 cost_cap 或采纳上述任一降级方案。" ) def charge(self, step: str, item: str, cost: float) -> None: """扣费,触发软预警/硬熔断.""" self.spent += cost self.records.append(CostRecord(step=step, item=item, cost=cost)) ratio = self.spent / self.cap if self.cap else 0 if ratio >= 1.0: self._persist() raise BudgetExceeded( f"累计成本 ¥{self.spent:.2f} 达到熔断上限 ¥{self.cap:.2f}({st ...[truncated 2360 chars]
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Findings (22)

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill claims a constrained workflow with fixed shot duration, limited concurrency, and last-frame chaining, yet analysis indicates those controls are not implemented while undeclared budget/cost persistence features are present. Missing advertised safety or operational constraints can mislead orchestrators into relying on protections that do not exist, while hidden persistence and cost-tracking introduce additional data-handling and side-effect risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
95% confidence
Finding

The skill claims a constrained workflow with fixed shot duration, limited concurrency, and last-frame chaining, yet analysis indicates those controls are not implemented while undeclared budget/cost persistence features are present. Missing advertised safety or operational constraints can mislead orchestrators into relying on protections that do not exist, while hidden persistence and cost-tracking introduce additional data-handling and side-effect risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The skill claims a constrained workflow with fixed shot duration, limited concurrency, and last-frame chaining, yet analysis indicates those controls are not implemented while undeclared budget/cost persistence features are present. Missing advertised safety or operational constraints can mislead orchestrators into relying on protections that do not exist, while hidden persistence and cost-tracking introduce additional data-handling and side-effect risks.

Content

No source excerpt is available for this finding.

Tp4

High
Category
MCP Tool Poisoning
Confidence
99% confidence
Finding

The skill claims a constrained workflow with fixed shot duration, limited concurrency, and last-frame chaining, yet analysis indicates those controls are not implemented while undeclared budget/cost persistence features are present. Missing advertised safety or operational constraints can mislead orchestrators into relying on protections that do not exist, while hidden persistence and cost-tracking introduce additional data-handling and side-effect risks.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
88% confidence
Finding

The skill declares operational behavior involving file access, environment usage, and likely network/API calls, but does not declare any tool scope such as permissions or allowed-tools. This creates an authorization ambiguity where an agent may invoke the skill with broader capabilities than users or reviewers expect, increasing the risk of unintended file or network access.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manifest describes a skill focused on turning storyboard keyframes into short videos using Seedance 2.0, but this shared client also exposes full image generation and text-to-speech functions. Those capabilities are not justified by the stated purpose of '分镜关键帧 → 短视频' and materially expand what the skill can do.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The code converts local image files to data URIs and includes them in HTTP requests to the external Ark API, which transmits user-provided file contents off the local system. There is no confirmation prompt, user-facing log, or warning in this file disclosing that referenced local files will be uploaded to a remote service.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/_shared/ark_api.py (reported line 66)May include surrounding context.

python
"size": size,
            "watermark": False,
        }
        resp = requests.post(
            f"{self.base_url}/images/generations",
            headers=self.headers,
            json=body,

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
82% confidence
Finding

The submit_video method includes reference_video and reference_audio values in the request body sent to the Ark service. Although this is part of the feature, this file provides no user-facing disclosure or confirmation that supplied media references may be transmitted to a third-party API.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/_shared/ark_api.py (reported line 144)May include surrounding context.

python
if seed is not None:
            body["seed"] = seed

        resp = requests.post(
            f"{self.base_url}/contents/generations/tasks",
            headers=self.headers,
            json=body,

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The manifest specifically scopes the skill to generating short videos from keyframes, mentioning first-frame chaining and video settings, but this module can also synthesize audio from arbitrary text and write it to disk. That is a distinct capability not implied by the declared trigger words or description.

Content

No source excerpt is available for this finding.

External Transmission

Medium
Category
Data Exfiltration
Confidence
80% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/_shared/ark_api.py (reported line 207)May include surrounding context.

python
if emotion:
            body["emotion"] = emotion

        resp = requests.post(
            f"{self.base_url}/audio/speech",
            headers=self.headers,
            json=body,

Tainted flow: 'resp' from requests.get (line 158, network input) → pathlib.Path.write_bytes (file write)

Medium
Category
Data Flow
Confidence
65% confidence
Finding

Data from a source is assigned to a variable that is later passed to a sink, creating a variable-mediated taint flow.

Content

Scanner excerpt · scripts/_shared/ark_api.py (reported line 216)May include surrounding context.

python
resp.raise_for_status()
        out = pathlib.Path(out_path)
        out.parent.mkdir(parents=True, exist_ok=True)
        out.write_bytes(resp.content)
        return out

    # ------------ 工具 ------------

External Transmission

Medium
Category
Data Exfiltration
Confidence
60% confidence
Finding

Data is being sent to an external URL. This could be legitimate telemetry or data exfiltration. Manual review is recommended.

Content

Scanner excerpt · scripts/_shared/config.py (reported line 35)May include surrounding context.

python
# API 端点(2026-04 核对)
ENDPOINTS = {
    "ark_base":      "https://ark.cn-beijing.volces.com/api/v3",
    "kling_base":    "https://api.klingai.com/v1",         # 官方;第三方 piapi.ai/fal.ai 亦可
    "suno_base":     "https://api.sunoapi.org",             # 第三方(Suno 无公开官方 API)
}

Natural-Language Policy Violations

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

This code file contains natural-language docstrings and runtime messages entirely in Chinese, including the module description, exception text, and warning output. Under the language/locale policy, forcing a specific language without user opt-in or a documented region-specific justification is a policy violation.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The code submits a local image file via client.submit_video using first_frame=str(frame_path), which implies transmitting user-provided content to an external service. While there are progress prints after submission, there is no explicit warning, confirmation, comment, or docstring here disclosing that local media will be uploaded to a remote API.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
76% confidence
Finding

The module-level natural-language documentation is entirely in Chinese, which can reflect an implicit language constraint without offering a user language choice or explaining a region-specific requirement. Under the language/locale policy, forced locale usage should be opt-in or clearly justified.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The download_video method fetches content from a remote URL and writes it to a local file path, creating directories as needed. This file operation is not accompanied by any confirmation prompt or visible disclosure in the function's documentation about writing to disk.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

The tts method sends text to a remote speech API and writes the binary audio response to out_path on disk. While the docstring mentions writing to out_path, there is no user-facing warning or confirmation about the external transmission of text content or the file write operation.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
87% confidence
Finding

The only natural-language description in this file is written in Chinese, which imposes a fixed language for users or maintainers reading the skill without indicating any language choice or locale-specific justification. This matches the policy category for language or locale constraints that are not clearly optional or justified.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
90% confidence
Finding

The module docstring and surrounding natural-language descriptions are entirely in Chinese, and the configuration hard-codes Chinese-language presets such as Chinese genres and voice IDs. For a general-purpose shared config, this reflects a locale constraint without any visible opt-in, alternative locale, or justification that the skill is region-specific.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

The script fetches a URL returned by the remote API and writes the response directly to disk without validating the URL origin, enforcing timeouts, checking content type, or limiting size. If the upstream service is compromised or returns attacker-controlled URLs, this can enable SSRF-like access from the host, unexpected local file creation, or resource exhaustion via oversized responses.

Content

No source excerpt is available for this finding.