Back to skill

Security audit

team-memory

Security checks for vulnerabilities and agentic risk

Overview

The skill’s automatic team notes, local storage, and Notion synchronization are disclosed and consistent with its stated purpose.

Install if you want relevant team conversations and personnel records saved automatically and synchronized to Notion. Restrict the integration to your intended private library and keep its token in the environment. Original text and revision history remain in backups and cloud packages after record deletion. Background synchronization can continue retrying while work is pending. Actual Notion access permissions were not verified.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Behavioral ASTexec() Call, eval() Call, Dynamic Import
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (6)

Tainted flow: 'headers' from os.environ.get (line 104, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
90% confidence
Finding

Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.

Content

Scanner excerpt · scripts/tm3/transport.py (reported line 108)May include surrounding context.

python
if body is not None: headers["Content-Type"]=content_type or "application/json"
            try:
                self.request_count+=1
                with urllib.request.urlopen(urllib.request.Request(url,data=body,headers=headers,method=method),timeout=45) as response:
                    result=json.loads(response.read() or b"{}")
                self.last_request=time.monotonic(); self.checkpoint(); return result
            except urllib.error.HTTPError as exc:

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

The synchronization workflow uploads memory records and history to Notion without a privacy warning.

Medium
Category
Output Handling
Confidence
88% confidence
Finding

The commit workflow instructs uploading incremental packages or full snapshots and updating person-specific Notion pages. These packages include current records, unpublished historical revisions, and new objects, but the document does not warn that potentially sensitive personal information will be stored with a third-party service.

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/tm3/sync.py (reported line 266)May include surrounding context.

python
options=dict(stdin=subprocess.DEVNULL,stdout=output,stderr=output,close_fds=True)
        if os.name=="nt": options["creationflags"]=subprocess.CREATE_NEW_PROCESS_GROUP|subprocess.DETACHED_PROCESS
        else: options["start_new_session"]=True
        subprocess.Popen(command,**options)
    return True

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/tm3/transport.py (reported line 98)May include surrounding context.

python
finally: self.inside_guard=False

    def request(self,method,path,data=None,raw=None,content_type=None,safe=False):
        if path.startswith("https://") and not path.startswith("https://api.notion.com/"):
            raise MemoryError("拒绝向其他服务发送Notion凭据")
        url=path if path.startswith("https://") else "https://api.notion.com"+path
        body=raw if raw is not None else encoded(data) if data is not None else None

External Transmission

Medium
Category
Data Exfiltration
Confidence
70% 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/tm3/transport.py (reported line 108)May include surrounding context.

python
if body is not None: headers["Content-Type"]=content_type or "application/json"
            try:
                self.request_count+=1
                with urllib.request.urlopen(urllib.request.Request(url,data=body,headers=headers,method=method),timeout=45) as response:
                    result=json.loads(response.read() or b"{}")
                self.last_request=time.monotonic(); self.checkpoint(); return result
            except urllib.error.HTTPError as exc: