Back to skill

Security audit

视频参考素材分工表

Security checks for vulnerabilities and agentic risk

Overview

The skill coherently organizes user-provided video references, with no evidence of hidden execution, data theft, or excessive authority.

Provide only reference materials you are authorized to share with your agent. If using the README installer, verify the publisher and prefer reviewed, pinned versions. The promotional-image helper is unnecessary for the reference-mapping workflow.

Vulnerability Patterns
  • 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
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • 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 (9)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied shell command performs skill installation, not the declared reference-mapping task. It invokes an external package runner and identifies an external skill source, introducing retrieval and installation behavior absent from the description. The installed skill's eventual behavior cannot be verified from this chunk.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
98% confidence
Finding

The supplied chunk is a promotional-image renderer, not an implementation of the declared reference-planning workflow. Its text mentions the advertised table, but it does not construct or validate one. Font discovery and local file access are ordinary implementation details; the material mismatch is the different output and primary behavior. No external downloading or network activity is shown.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

The supplied command performs a skill-installation operation, not the declared reference-mapping workflow. Although installation could be a setup step for a larger implementation, no such implementation is supplied, and the behavior of the installed skill cannot be verified from this chunk.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
91% confidence
Finding

The installation command uses npx skills without a version pin, allowing it to fetch and execute the currently resolved CLI package with the invoking user's privileges. A compromised upstream release could therefore execute malicious code; this is a supply-chain reproducibility weakness, not evidence of an MCP server or deliberate malicious behavior. Its placement in an optional, manually invoked installation example reduces immediate exposure.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
87% confidence
Finding

The installation command invokes npx skills without a fixed version, potentially downloading and executing a mutable npm package with the installing user's privileges. A compromised future release could therefore execute malicious code, although this README contains no evidence of malicious intent or an actual compromise. The finding's MCP-server characterization is inaccurate: this is a CLI installation supply-chain risk, and its placement in installation documentation makes exploitation dependent on someone running the command.

Content

No source excerpt is available for this finding.

Skill declares no tool scope ('permissions' or 'allowed-tools') but code capabilities were detected: file_read, shell.

Medium
Category
MCP Least Privilege
Confidence
70% confidence
Finding

Without declared permissions the skill's intent is opaque and cannot be validated.

Content

No source excerpt is available for this finding.

subprocess module call

Medium
Category
Dangerous Code Execution
Confidence
70% confidence
Finding

subprocess module calls execute external commands. Without careful input validation, this enables command injection.

Content

Scanner excerpt · scripts/render_promo_fallback.py (reported line 10)May include surrounding context.

python
if len(sys.argv) != 3:
    raise SystemExit("usage: render_promo_fallback.py INPUT OUTPUT")
font = subprocess.check_output(["fc-match", "-f", "%{file}", "PingFang SC"], text=True).strip()
if not font or not Path(font).is_file():
    raise SystemExit("no usable Chinese font discovered by fontconfig")
source = Image.open(sys.argv[1]).convert("RGB")

Scope Creep

Low
Category
Excessive Agency
Confidence
70% confidence
Finding

Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.

Content

Scanner excerpt · LICENSE (reported line 12)May include surrounding context.

text
permit persons to whom the Software is furnished to do so.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A
PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT
HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE

The script generates a promotional image, a capability outside the stated reference-role mapping purpose.

Low
Category
Excessive Agency
Confidence
88% confidence
Finding

L13–L24 resize an input image, overlay fixed marketing slogans, and save a PNG. This produces promotional artwork rather than organizing reference materials into role assignments or identifying conflicts and missing materials, and the manifest does not declare promotional-image generation.

Content

No source excerpt is available for this finding.