Back to skill

Security audit

品牌 GEO 知识库整理

Security checks for vulnerabilities and agentic risk

Overview

The skill coherently organizes selected local documents, with an optional, disclosed publishing workflow that users should review before sharing content.

The output retains complete source copies, so protect it like the originals. Before using optional publishing, check for confidential information and specify the approved articles, accounts, and schedule. Review separate aimeGeo software before installing it. The supplied malicious signal is unsupported by the inspected artifacts; the supplied A.I.G report contains no findings.

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
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Unvalidated Output Injection

High
Category
Output Handling
Confidence
95% confidence
Finding

Model output is used without validation or sanitization. Unvalidated output injected into downstream contexts (SQL, shell, HTML) enables injection attacks and arbitrary code execution.

Content

Scanner excerpt · scripts/self_test.py (reported line 33)May include surrounding context.

python
self.payload = json.loads((self.inputs / "input.json").read_text(encoding="utf-8"))

    def run_build(self, name="input.json", success=True, output=None):
        result = subprocess.run(
            [sys.executable, str(ROOT / "scripts" / "knowledge_base.py"), "build",
             "--input", str(self.inputs / name), "--output", str(output or self.output)],
            cwd=self.work, capture_output=True, encoding="utf-8",

The continuation request directs use of existing publishing accounts and schedules without warning that source-derived content may be disclosed externally.

Medium
Category
Output Handling
Confidence
90% confidence
Finding

The copyable request instructs the agent to read the knowledge base, create content, and continue through existing accounts and schedules; the surrounding workflow identifies these as publishing accounts. Unlike a tool listing, this is an actionable instruction, but it provides no warning that potentially nonpublic brand facts could become externally visible.

Content

No source excerpt is available for this finding.

The copyable workflow directs publication through existing accounts without warning that local knowledge-base content may become public.

Medium
Category
Output Handling
Confidence
91% confidence
Finding

This is an actionable instruction to create articles from local materials and arrange publication, rather than a description of another tool's capabilities. The README does not warn users to check those materials for confidential or non-public information before publication; restricting content to supported facts does not establish permission to disclose it.

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, file_write, 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/self_test.py (reported line 33)May include surrounding context.

python
self.payload = json.loads((self.inputs / "input.json").read_text(encoding="utf-8"))

    def run_build(self, name="input.json", success=True, output=None):
        result = subprocess.run(
            [sys.executable, str(ROOT / "scripts" / "knowledge_base.py"), "build",
             "--input", str(self.inputs / name), "--output", str(output or self.output)],
            cwd=self.work, capture_output=True, encoding="utf-8",

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

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 · NOTICE.md (reported line 26)May include surrounding context.

md
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