ClawHub
Back to skill

Security audit

Priority Coach Clawhub

Security checks across malware telemetry and agentic risk

Overview

This is a local priority-coaching skill that can save personal planning notes on the user's machine, with no evidence of upload, hidden execution, or unrelated access.

Install only if you are comfortable with a coaching skill asking about priorities, stressors, and regrets. Saved records stay local in ~/.workbuddy/priority-coach/records.json according to the artifact, but they can include sensitive reflections, so use delete or avoid saving if you do not want a local history.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
92% confidence
Finding
The skill instructs the agent to read and write local files via `scripts/record.py` and `~/.workbuddy/priority-coach/records.json`, but no permissions are explicitly declared. This creates a capability/consent mismatch: a seemingly benign coaching skill can persist sensitive personal reflections to disk, retrieve prior entries, and delete records without a clear permission boundary, increasing privacy and integrity risk.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The trigger phrases are broad, everyday natural-language requests such as feeling busy, wanting a plan, or needing to clarify priorities. In an assistant platform that auto-invokes skills from conversational text, this can cause unintended activation, misrouting user requests, and unnecessary exposure of personal planning context to the skill when the user did not explicitly intend to use it.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The activation examples are broad enough to match ordinary planning, mood, and self-improvement requests, which can cause the skill to be invoked unexpectedly in unrelated contexts. Because this skill includes local persistence of sensitive self-reports, overbroad triggering increases the chance that users disclose personal data or that file operations occur when they did not intend to enter this workflow.

Vague Triggers

Medium
Confidence
80% confidence
Finding
The 'when to use' section defines activation in very general emotional and planning terms, making the skill prone to intercept broad categories of everyday conversation. In this skill's context, that is more dangerous than a pure text-only helper because it can steer users into a structured intake about stress, priorities, and regret, then potentially save that sensitive information locally.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.