Back to skill

Security audit

Investment Analysis Workflow

Security checks for vulnerabilities and agentic risk

Overview

This skill is coherent as an investment-analysis workflow, but it can automatically change persistent investment files and gives ambiguous high-stakes trading guidance.

Install only if you want an agent to maintain and modify your local investment research files. Before use, require explicit confirmation before any edit or rewrite, keep backups or version control, and treat any profit forecasts, PE judgments, or buy/sell ranges as non-professional investment analysis rather than trading advice.

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
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (6)

Ae4

Medium
Category
analysis-evasion
Confidence
80% confidence
Finding

Suspicious Unicode normalization or mixed-script content

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The skill is described as a workflow manager, but it expands into producing substantive weekly investment research with firm judgments, profit forecasts, and portfolio-maintenance outputs. Scope drift like this is risky because a user invoking a benign-sounding workflow skill may trigger high-stakes financial analysis and decision-support behavior beyond what they reasonably expected.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The trigger phrases are very broad and overlap with ordinary requests like 'analyze new material' or 'update stock pool.' Broad activation criteria increase the chance that this skill auto-runs in situations where the user did not intend financial-workflow execution, file mutation, or cross-skill invocation, causing overreach and unintended actions.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
96% confidence
Finding

The skill directs the agent to append, update, and write multiple repository files as a mandatory step, but it does not require explicit user confirmation or warn that persistent changes will be made. In an agent environment, this creates a real integrity risk: ordinary analysis requests can silently mutate tracked files, archives, and evaluation artifacts.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The instruction to rewrite the consolidated evaluation file without preserving the old version or warning about overwrite behavior is destructive by design. If followed automatically, it can erase historical assessments, reduce auditability, and permanently replace prior judgments with newer output that may be incomplete or wrong.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

This section gives conflicting instructions: it says not to provide buy/sell price guidance, then explicitly permits buy/sell ranges. Contradictory policy in an autonomous skill is dangerous because downstream agents may choose the more permissive branch and generate regulated or high-risk financial advice that the surrounding workflow appears to forbid.

Content

No source excerpt is available for this finding.