ClawHub
Back to skill

Security audit

Pantry Tracker

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent pantry-tracking purpose, but it needs review because it stores grocery/email-derived data in Supabase with weak access-control guidance and some commands mutate records automatically.

Review before installing. Use this only with a Supabase project that has explicit per-user row-level security, restricted grants, and authenticated user tokens. Treat grocery emails and order IDs as personal data, avoid recurring email ingestion until you have confirmed consent, and prefer dry-run or confirmation behavior before commands mutate pantry records.

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)

T05 · Unauthorized Access and Privilege Escalation

Error
Location
references/supabase-schema.sql:4
Finding

Pantry Records Lack Row-Level Access Control

Content
View full analysis
Remediation
View remediation

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/pantry.py:42
Finding

Unencoded User Input Is Embedded in Supabase REST Filters

Content
View full analysis
Remediation
View remediation

SkillSpector

Vulnerability Patterns
  • 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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Findings (6)

Tainted flow: 'req' from os.environ.get (line 44, 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/pantry.py (reported line 46)May include surrounding context.

python
body = json.dumps(data).encode() if data else None
    req = Request(url, data=body, headers=_headers(), method=method)
    try:
        with urlopen(req) as resp:
            raw = resp.read().decode()
            return json.loads(raw) if raw.strip() else []
    except HTTPError as e:

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
87% confidence
Finding

The skill declares environment and network-backed behavior but does not define explicit tool scoping such as allowed tools or permissions. That creates ambiguity about what the agent may invoke at runtime and can lead to overbroad access, especially because the workflow includes external email parsing and Supabase writes.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
86% confidence
Finding

The invocation description is broad enough that the skill could be selected for many ordinary food- or pantry-related prompts without clear user intent to store data remotely. In context, that increases the chance the agent performs persistence actions or triggers email-driven ingestion when the user only wanted advice or a simple answer.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill explains that grocery order emails are parsed and pantry items are stored in Supabase, but it does not prominently warn users that purchase-derived data and pantry records will be persisted to a third-party backend. This can cause unintentional disclosure of shopping habits, household consumption patterns, and order metadata if the user assumes processing is local or ephemeral.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The refresh command performs remote PATCH operations that modify pantry item status in bulk without any dry-run, confirmation, or explicit warning at execution time. In an agent setting, this can cause unintended state changes from an ambiguous or read-only-seeming request, reducing auditability and potentially causing data integrity issues.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The summary command appears informational but implicitly calls refresh(), which performs remote writes before generating output. Hidden side effects are dangerous in agent workflows because a user or orchestrator may invoke summary expecting a safe read and instead trigger database mutations.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.