ClawHub
Back to skill

Security audit

project-brainstorming

Security checks across malware telemetry and agentic risk

Overview

This is a project-brainstorming skill, but it automatically saves and forwards project details into other agent/LLM workflow steps without a clear confirmation checkpoint.

Install only if you are comfortable with project brainstorming content being saved locally and automatically passed into other Attune workflow steps and subagents. Avoid using it for confidential business, customer, legal, security, or proprietary planning unless you have reviewed the downstream skills and are prepared to manage .attune/brainstorm-session.json and docs/project-brief.md.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list includes generic terms like "planning" and "requirements" that are common in normal developer conversation and can cause the skill to activate when the user did not intend to invoke it. Because this skill performs automatic downstream actions and context forwarding, accidental invocation increases the chance of unintended data disclosure and workflow execution.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill states that War Room integration is required and that all brainstorming context is automatically passed onward, but the top-level description does not clearly warn users up front that another skill/LLM will receive their content. This creates a transparency and consent failure: users may share sensitive project or business details during brainstorming without realizing they will be forwarded automatically.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill persists session data to `.attune/brainstorm-session.json`, including problem statements, stakeholders, constraints, and approaches, without a clear warning in the main description. This can silently write sensitive planning or proprietary information to disk where it may be retained, committed, or accessed by other tools.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- `docs/project-brief.md` does not exist or is empty (phase failed)
- User explicitly requests to stop after brainstorming

**Do NOT prompt the user for confirmation** — this is a lightweight checkpoint, not an interactive gate. The user can always interrupt if needed.


### Phase 6.5: Spec Review Gate
Confidence
92% confidence
Finding
The instruction to auto-continue and specifically "Do NOT prompt the user for confirmation" removes an important consent checkpoint before launching the next phase. In this skill, that is more dangerous because the workflow also includes file writes, review loops, and delegation to other skills/subagents, so an accidental or misunderstood run can cascade into additional actions without explicit approval.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.