Security audit

Alphagbm Theme Research

Security checks for vulnerabilities and agentic risk

Overview

This skill coherently manages AlphaGBM investment themes through a disclosed remote API, with some ordinary caution needed around external data sharing and hard deletion.

Install only if you are comfortable sending your theme names, tickers, descriptions, and news keywords to AlphaGBM using your API key. Before any update or hard delete, make sure the agent has resolved the intended theme and confirms the name or ID, especially when your request uses generic words like theme or basket.

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

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (3)

Tool Parameter Abuse

High

Category: Tool Misuse
Confidence: 84% confidence
Finding: The skill exposes a hard-delete endpoint for themes, and the documented workflow relies on resolving IDs then issuing deletion without any explicit confirmation or ownership-safety UX beyond server-side authorization. In a conversational system, broad triggers or ambiguous references can lead to destructive actions being invoked on the wrong resource, causing irreversible loss of user data.
Content: Scanner excerpt · SKILL.md (reported line 103)May include surrounding context.
5. Delete theme (by id)

text
DELETE /api/research/themes/<THEME_ID>

Hard-delete. Doesn't affect the underlying company profiles.

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 87% confidence
Finding: The trigger guidance is broad enough to match generic words like 'theme', 'basket', or '板块', which can cause the skill to activate when the user did not intend to manage AlphaGBM themes. Because the skill can perform state-changing actions against a remote API, unintended invocation increases the chance of accidental data disclosure or unwanted theme modifications.
Content: No source excerpt is available for this finding.

Missing User Warnings