Back to skill

Security audit

多账号发布交接与回执

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently prepares local publishing handoff packages, with optional publishing limited to specifically authorized accounts and manuscripts.

Protect the package because it contains manuscript copies and account history. Review optional publishing tools separately and authorize specific accounts and content. Receipt labels depend on evidence supplied by the user or executor, so verify actual platform results. Preserve the ledger: duplicate-submission protection does not extend across independent packages.

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
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • 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
Findings (10)

Description-Behavior Mismatch

High
Category
MCP Tool Poisoning
Confidence
97% confidence
Finding

交接包整理、原稿保留、历史维护和防重投状态控制与描述一致,但描述还承诺了外部执行、回读和引用检测,所供代码均未实现。代码没有网络请求、浏览器操作或 aimeGeo 服务集成;即使回执标记为 observed,其真实性也依赖外部输入,而非脚本独立验证。因此存在能力范围不匹配,但未发现额外的未声明网络访问或发布行为。

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, network, 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 34)May include surrounding context.

python
self.tick = datetime.now(timezone.utc)

    def call(self, *args, error=None, cli=CLI):
        result = subprocess.run([sys.executable, str(cli), *map(str, args)], capture_output=True, text=True)
        self.assertEqual(result.returncode, 2 if error else 0, result.stdout + result.stderr)
        data = json.loads(result.stderr if error else result.stdout)
        if error:

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/self_test.py (reported line 269)May include surrounding context.

python
self.record(self.receipt("pre_filled", mode="user_report"))
        self.begin(error="READBACK_REQUIRED")
        self.record(self.receipt("pre_filled"))
        self.begin(error="SUBMIT_NOT_AUTHORIZED", auth=False)
        result = self.begin()
        self.assertEqual(result["state"], "unknown")
        self.assertEqual(len(self.target()["intents"]), 1)

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/self_test.py (reported line 390)May include surrounding context.

python
self.assertEqual(self.target()["intents"], history["existing_intents"])
        self.assertEqual(self.target()["state"], "pre_filled")
        self.assertEqual(self.target()["authorization"]["action"], "prepare")
        self.begin(auth=False, error="SUBMIT_NOT_AUTHORIZED")
        fresh = self.begin()
        self.assertNotEqual(fresh["intent_id"], old_intent["intent_id"])
        self.begin(error="READBACK_REQUIRED")

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/self_test.py (reported line 482)May include surrounding context.

python
self.prepare(history)
        self.assertEqual(self.target()["state"], "pre_filled")
        self.assertEqual(self.target()["intents"], history["existing_intents"])
        self.begin(auth=False, error="SUBMIT_NOT_AUTHORIZED")
        self.assertEqual(self.target("zhihu")["state"], "under_review")

    def test_import_submitted_receipt_with_intent_keeps_permanent_protection(self):

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/self_test.py (reported line 321)May include surrounding context.

python
self.record(self.receipt("failed", submission_absent=True, intent_id=first["intent_id"]))
        self.begin(error="READBACK_REQUIRED")
        self.record(self.receipt("pre_filled"))
        second = self.begin(auth=False)  # Saved authorization remains scoped; do not ask again.
        self.assertNotEqual(first["intent_id"], second["intent_id"])
        self.assertEqual(len(self.target()["intents"]), 2)
        self.record(self.receipt("failed", submission_absent=True, intent_id=first["intent_id"]), error="INTENT_MISMATCH")

Unsafe Defaults

Medium
Category
Tool Misuse
Confidence
75% confidence
Finding

Tool defaults are unsafe or overly permissive (e.g. disabled TLS verification, no authentication, world-writable permissions). Unsafe defaults widen the attack surface.

Content

Scanner excerpt · scripts/self_test.py (reported line 685)May include surrounding context.

python
self.record(self.receipt("failed", submission_absent=True), error="INTENT_MISMATCH")
        self.record(self.receipt("failed", submission_absent=True, intent_id=first["intent_id"]))
        self.record(self.receipt("pre_filled"))
        self.begin(auth=False)
        self.assertEqual(len(self.target()["intents"]), 2)

    def test_untrusted_late_failure_or_simulation_cannot_clear_unknown(self):

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 843)May include surrounding context.

python
self.record(self.receipt("pre_filled"))
        args = [sys.executable, str(CLI), "intent", "--pack", str(self.pack), "--target", self.target()["id"],
                "--authorization", str(self.grant())]
        processes = [subprocess.Popen(args, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True) for _ in range(2)]
        results = [(p, p.communicate(timeout=20)) for p in processes]
        self.assertEqual(sorted(p.returncode for p, _ in results), [0, 2], results)
        self.assertEqual(len(self.target()["intents"]), 1)

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