Back to skill

Security audit

Stock Analysis

Security checks for vulnerabilities and agentic risk

Overview

The skill uses CellCog for financial research as advertised, with ordinary privacy, dependency, and background-notification considerations.

Prompts are processed by CellCog. Use public or redacted financial information and review its data policies before sharing confidential material. Review and pin dependencies where practical, protect your API key, and expect service-credit charges and background notifications.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Output HandlingUnvalidated Output Injection, Cross-Context Output, Unbounded Output
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

The workflow sends task prompts to CellCog without warning that sensitive financial information will leave the user's environment.

Medium
Category
Output Handling
Confidence
96% confidence
Finding

The examples send task prompts to an external service, while the surrounding workflow encourages inclusion of personal financial details and potentially confidential business information without an explicit disclosure or data-minimization safeguard. External processing is an intended feature rather than covert exfiltration, but the financial context makes inadvertent disclosure more consequential than ordinary public-stock research.

Content

No source excerpt is available for this finding.

MCP server referenced without pinned version: 'npx skills'.

Medium
Category
MCP Rug Pull
Confidence
85% confidence
Finding

The setup command invokes an unpinned npm executable and installs skill content without an immutable revision, creating a supply-chain exposure if an upstream release or repository is compromised. This is an installation workflow, not an MCP server reference as the detector claims; the risk is execution of unreviewed upstream code or installation of altered agent instructions, not evidence of an existing malicious payload.

Content

No source excerpt is available for this finding.