ClawHub
Back to skill

Security audit

Alphagbm Health Check

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent AlphaGBM health-check helper, but it exposes destructive profile-deletion guidance without a clear confirmation requirement.

Install only if you want an agent to access your AlphaGBM research workspace using ALPHAGBM_API_KEY. Treat health reports as read-only unless you explicitly ask for a specific follow-up action, and do not allow archive/delete actions unless the agent repeats the exact ticker and you confirm the deletion separately.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Confidence
86% confidence
Finding
The trigger list includes broad phrases like "generate a report" and generic health/audit wording that can match routine conversation outside the intended research-KB context. Overbroad activation can cause unintended invocation of the skill, exposing workspace metadata or prompting downstream actions the user did not explicitly request.

Vague Triggers

Medium
Confidence
83% confidence
Finding
The when-to-use section contains ambiguous examples such as "audit," "health check," and "什么需要更新" without firm boundaries tying them to the AlphaGBM research workspace. This increases the chance of unintended routing and can cause the agent to apply the skill in contexts where the user did not intend KB diagnostics.

Missing User Warnings

Medium
Confidence
88% confidence
Finding
The skill describes follow-on recommendation actions including archive and delete behavior but does not prominently require a user warning or explicit confirmation before destructive operations. In an agentic workflow, this can normalize deletion as a routine remediation step and lead to accidental loss of research profiles or orphaned content.

Tool Parameter Abuse

High
Category
Tool Misuse
Content
- `refresh` — call `POST /api/research/profiles/<ticker>/refresh`
- `review_thesis` — surface the thesis + updated fundamentals; user decides to edit/close
- `archive` — `DELETE /api/research/profiles/<ticker>` or delete orphan
- `create_thesis` — profile exists but has no thesis; prompt user to write one

Skills that execute these actions: `alphagbm-company-profile`, `alphagbm-investment-thesis`.
Confidence
93% confidence
Finding
DELETE /api/research/profiles/<ticker>`

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.