Back to skill

Security audit

passnote-skill

Security checks for vulnerabilities and agentic risk

Overview

This PassNote skill does what it claims, but it handles secrets and an API token with weak endpoint safety and incomplete disclosure.

Review before installing. Only configure PASSNOTE_API_URL to a PassNote server you trust, prefer HTTPS, avoid using this for high-value passwords or regulated data unless you control the server and token scope, and rotate the token if it may have been sent to an untrusted or plaintext endpoint.

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

Error
Location
scripts/create_memo.py:19
Finding

Sensitive memo content and bearer token may be transmitted over an insecure connection

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/create_memo.py:44
Finding

Viewing URL is derived using non-structural hostname replacement

Content
View full analysis
Remediation
View remediation
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Rogue AgentSelf-Modification, Session Persistence
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Taint TrackingDirect Taint Flow, Variable-Mediated Taint Flow, Credential Exfiltration Chain
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
Findings (9)

Tainted flow: 'req' from os.environ.get (line 33, credential/environment) → urllib.request.urlopen (network output)

Critical
Category
Data Flow
Confidence
93% confidence
Finding

The request destination is derived from PASSNOTE_API_URL, an environment variable, while the Authorization bearer token is attached to every request. If an attacker can influence that environment variable, the script will send the secret token and memo content to an attacker-controlled endpoint, creating a straightforward credential exfiltration and sensitive-data disclosure path. The default also permits plain HTTP, which increases interception risk.

Content

Scanner excerpt · scripts/create_memo.py (reported line 38)May include surrounding context.

python
req.add_header("Authorization", f"Bearer {api_token}")
    
    try:
        response = urllib.request.urlopen(req)
        response_data = json.loads(response.read().decode("utf-8"))
        
        if response_data.get("code") == 0 and "data" in response_data:

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · README.md (reported line 3)May include surrounding context.

md
# OpenClaw PassNote Skill

This skill allows the OpenClaw assistant to create and manage secure, self-destructing memos using your PassNote service.

## Installation and Configuration

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The README encourages users to provide sensitive memo content to the assistant and says it will run the underlying script, but it does not clearly warn that the memo text is sent to the external PassNote service. This can mislead users into sharing passwords or other secrets under the assumption the content stays local, increasing the risk of unintended disclosure to a remote system.

Content

No source excerpt is available for this finding.

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding

The skill declares environment-variable and network-dependent behavior but does not define an explicit tool scope such as allowed tools or permissions. This creates ambiguity about what the skill is permitted to access and increases the chance of over-broad execution in agent environments, especially since it handles sensitive memo contents and API credentials.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · SKILL.md (reported line 3)May include surrounding context.

md
---
name: passnote
description: Create and manage disposable memos using PassNote. Share secure, auto-destructing notes with others.
env:
  - PASSNOTE_API_URL
  - PASSNOTE_API_TOKEN

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
89% confidence
Finding

The activation text is broad enough to match ordinary requests involving notes, memos, or messages, which could trigger the skill in situations where the user did not intend to send content to an external service. Because the skill transmits potentially sensitive text to PassNote, over-triggering raises the risk of unintended disclosure.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill does not warn users that memo contents will be transmitted to an external PassNote service. Users may reasonably assume note handling is local, so the lack of disclosure can cause accidental sharing of secrets, credentials, or regulated data with a third party.

Content

No source excerpt is available for this finding.

Session Persistence

Medium
Category
Rogue Agent
Confidence
60% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

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

python
import urllib.error

def main():
    parser = argparse.ArgumentParser(description="Create a PassNote memo")
    parser.add_argument("content", type=str, help="The content of the memo")
    parser.add_argument("--expire-hours", type=int, default=24, help="Hours until the memo expires (1-48)")

Session Persistence

Medium
Category
Rogue Agent
Confidence
75% confidence
Finding

Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Content

Scanner excerpt · scripts/create_memo.py (reported line 45)May include surrounding context.

python
memo_data = response_data["data"]
            pass_key = memo_data.get("pass_key")
            
            frontend_url = api_url.replace("api.", "") # simplistic guess for frontend URL
            if "localhost" in frontend_url or "127.0.0.1" in frontend_url:
                frontend_url = "http://localhost:5173"