ClawHub
Back to skill

Security audit

Alphagbm Health Check

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent AlphaGBM research health-check skill, but users should be careful with broad trigger phrases and any archive/delete recommendation.

Install this only if you intend to let the agent read and generate AlphaGBM research health reports with your API key. Treat broad requests like 'generate a report' as ambiguous, and require a separate confirmation before any refresh, archive, or delete action on a ticker.

A.I.G

Vulnerability Patterns
  • 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
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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)

Tool Parameter Abuse

High
Category
Tool Misuse
Confidence
88% confidence
Finding

The skill propagates a raw path template containing <ticker> into a DELETE endpoint without any stated validation, allowlisting, or confirmation controls. In an agentic system, untrusted or mis-resolved ticker input could target unintended resources, and because the operation is destructive, mistakes can directly delete user research profiles.

Content

Scanner excerpt · SKILL.md (reported line 121)May include surrounding context.

md
- `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`.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The trigger list includes broad natural-language phrases like "generate a report" and generic health-check wording that can match routine conversation outside the intended AlphaGBM research domain. This can cause unintended skill activation and subsequent API calls against a user's research workspace, exposing private workspace state or initiating actions the user did not clearly request.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
92% confidence
Finding

The "When to use" section authorizes activation on vague phrases like "audit," "health check," and "什么需要更新," which are common conversational terms with no hard boundary to this specific service. In an agent environment, this increases the chance the skill runs in the wrong context and fetches or acts on AlphaGBM data when the user meant something else.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill presents destructive archive/delete actions as ordinary recommendations and workflow steps, but does not require an explicit warning or confirmation about permanent data removal. A user or downstream agent following the recommendation could delete research profiles or orphan pages without understanding the data-loss consequences.

Content

No source excerpt is available for this finding.