ClawHub
Back to skill

Security audit

gmail-checker

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent Gmail helper, but it needs Review because it can send, read, and modify email using powerful Gmail credentials without enough confirmation, scoping, or credential-safety guidance.

Install only if you are comfortable giving the skill access to send, read, and modify your Gmail. Use the narrowest OAuth scopes you can, avoid long-lived refresh tokens unless necessary, keep tokens out of prompts and logs, and require manual confirmation before any send, reply, label change, or bulk mailbox action.

A.I.G

Vulnerability Patterns
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs
  • 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
Findings (2)

T09 · Insecure Skill Coding Practices

Warning
Location
SKILL.md:51
Finding

OAuth Credentials Exposed Through Command-Line Arguments

Content
View full analysis
Remediation
View remediation

T05 · Unauthorized Access and Privilege Escalation

Note
Location
SKILL.md:270
Finding

Overly Broad Google Network Policy Exceeds Runtime Requirements

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (14)

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 21)May include surrounding context.

md
## Setup

### Option 1: OAuth2 Access Token (recommended)

1. Create OAuth2 credentials at https://console.cloud.google.com/apis/credentials
2. Enable the Gmail API at https://console.cloud.google.com/apis/library/gmail.googleapis.com

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 25)May include surrounding context.

md
1. Create OAuth2 credentials at https://console.cloud.google.com/apis/credentials
2. Enable the Gmail API at https://console.cloud.google.com/apis/library/gmail.googleapis.com
3. Obtain an access token via OAuth2 flow with scopes:
   - `https://www.googleapis.com/auth/gmail.send`
   - `https://www.googleapis.com/auth/gmail.readonly`
   - `https://www.googleapis.com/auth/gmail.modify`

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 53)May include surrounding context.

md
1. Create OAuth2 credentials at https://console.cloud.google.com/apis/credentials
2. Enable the Gmail API at https://console.cloud.google.com/apis/library/gmail.googleapis.com
3. Obtain an access token via OAuth2 flow with scopes:
   - `https://www.googleapis.com/auth/gmail.send`
   - `https://www.googleapis.com/auth/gmail.readonly`
   - `https://www.googleapis.com/auth/gmail.modify`

Credential Access

High
Category
Privilege Escalation
Confidence
70% confidence
Finding

Code accesses credential files (SSH keys, AWS credentials, etc.). This could indicate credential theft attempts.

Content

Scanner excerpt · SKILL.md (reported line 303)May include surrounding context.

md
- Gmail API rate limit: 250 quota units per second per user.
- Sending has a daily limit of 2,000 messages for Google Workspace, 500 for free Gmail.
- Access tokens expire after ~1 hour. Use refresh token flow for long-running sessions.
- Attachments require multipart MIME encoding — for large files, use the resumable upload endpoint.

## Examples

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill is user-invocable and performs high-impact actions such as sending mail, replying, drafting, and modifying labels, yet it lacks an upfront requirement for confirmation before state-changing operations. That makes accidental or socially engineered email actions more likely, especially when prompts are ambiguous or maliciously crafted.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The setup section instructs users to provide powerful Gmail credentials and refresh tokens but does not warn about their sensitivity, scope minimization, storage handling, or the risks of long-lived tokens. This increases the chance of insecure deployment and overprivileged access to a mailbox.

Content

No source excerpt is available for this finding.

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 · SKILL.md (reported line 47)May include surrounding context.

All API calls use Bearer auth:

bash
curl -s -H "Authorization: Bearer $GMAIL_ACCESS_TOKEN" \
  "https://gmail.googleapis.com/gmail/v1/users/me/..."

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill expands from Gmail operations into arbitrary web searching and page fetching, which is outside the declared Gmail-only scope. That creates unnecessary data egress and prompt-surface expansion: email content or user-provided topics could be sent to third-party sites, and fetched content could influence generated emails in unsafe ways.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

A Gmail management skill does not need general web browsing to fulfill its core purpose, so this capability materially broadens attack surface without justification. It enables unintended external transmission of user prompts or email-derived content to arbitrary websites and introduces untrusted content into downstream email composition.

Content

No source excerpt is available for this finding.

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 · SKILL.md (reported line 112)May include surrounding context.

md
ENCODED=$(echo -n "$RAW_MESSAGE" | base64 -w 0 | tr '+/' '-_' | tr -d '=')

# Send
curl -s -X POST \
  -H "Authorization: Bearer $GMAIL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  "https://gmail.googleapis.com/gmail/v1/users/me/messages/send" \

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 · SKILL.md (reported line 205)May include surrounding context.

md
ENCODED=$(echo -n "$RAW_MESSAGE" | base64 -w 0 | tr '+/' '-_' | tr -d '=')

# Send
curl -s -X POST \
  -H "Authorization: Bearer $GMAIL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  "https://gmail.googleapis.com/gmail/v1/users/me/messages/send" \

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 · SKILL.md (reported line 221)May include surrounding context.

bash
ENCODED=$(echo -n "$RAW_MESSAGE" | base64 -w 0 | tr '+/' '-_' | tr -d '=')

curl -s -X POST \
  -H "Authorization: Bearer $GMAIL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  "https://gmail.googleapis.com/gmail/v1/users/me/drafts" \

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 · SKILL.md (reported line 240)May include surrounding context.

md
"https://gmail.googleapis.com/gmail/v1/users/me/labels" | jq '.labels[] | {name, id}'

# Add label to a message
curl -s -X POST \
  -H "Authorization: Bearer $GMAIL_ACCESS_TOKEN" \
  -H "Content-Type: application/json" \
  "https://gmail.googleapis.com/gmail/v1/users/me/messages/{MESSAGE_ID}/modify" \

Description-Behavior Mismatch

Low
Category
Not specified by scanner
Confidence
85% confidence
Finding

The proposed network policy grants broad GET/POST access to accounts.google.com, but the documented flows only require Gmail API and token refresh endpoints. Overbroad egress permissions increase the chance of unintended authentication interactions, data leakage, or abuse if the skill or surrounding agent is prompted to access unrelated Google account pages.

Content

No source excerpt is available for this finding.