Security audit

Rivalwatch

Security checks for vulnerabilities and agentic risk

Overview

This skill is a local competitor-intelligence logging utility, not an exfiltrating or destructive tool, but its analysis/reporting language is stronger than what the script actually does.

Install only if you want a local logbook for competitor notes. Do not enter confidential strategy, pricing, or customer information unless you are comfortable with it being stored under ~/.local/share/rivalwatch/ and potentially exported to local files.

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

Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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 (8)

Tp4

High

Category: MCP Tool Poisoning
Confidence: 99% confidence
Finding: The skill is presented as a competitor-analysis generator, but the documented behavior includes persistent logging of arbitrary user inputs, local search across stored entries, export of accumulated data, and filesystem/status reporting. In a competitive-intelligence context, inputs are likely to contain sensitive business information, so undisclosed retention and export capabilities materially expand data exposure beyond the declared purpose.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

High

Category: Not specified by scanner
Confidence: 98% confidence
Finding: The advertised skill is for competitor analysis, but the implementation is a generic local logging utility with unrelated commands and no analysis logic. This mismatch is dangerous because it expands the tool's effective capability surface beyond user expectations, making it easier to collect or retain data under a misleading description.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The manifest describes a tool for analysis and report generation, but the implementation described in the file is primarily a local logging and data-management CLI. This discrepancy can mislead users into supplying sensitive competitor or market data under the assumption it will be transiently processed rather than automatically stored and managed on disk.
Content: No source excerpt is available for this finding.

Intent-Code Divergence

Medium

Category: Not specified by scanner
Confidence: 93% confidence
Finding: The command documentation uses action-oriented language such as run, analyze, generate, compare, and report, but the described actual behavior is only to show recent log entries or append timestamped input. That overstatement is security-relevant because it obscures the real data flow: user-supplied content is being persistently recorded rather than meaningfully processed as expected.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The skill does not provide an explicit warning that supplied inputs are automatically written to local log files. Because competitor-analysis workflows often involve confidential strategy, pricing, and market intelligence, silent persistence increases the risk of unintended retention, later disclosure through search/export, and leakage to other local users or backup systems.
Content: No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium

Category: Not specified by scanner
Confidence: 97% confidence
Finding: The script creates a persistent data directory and logs activity/history, which is not disclosed by the skill description. In this context, arbitrary user-supplied competitor research inputs may contain sensitive business information, so silent retention on disk creates privacy and data-handling risk.
Content: No source excerpt is available for this finding.

Missing User Warnings

Medium

Category: Not specified by scanner
Confidence: 96% confidence
Finding: User input is written directly to log files without any upfront notice in help text or comments, so users may unknowingly persist sensitive competitor-analysis queries and notes locally. In a business-analysis context, these inputs can include confidential market strategy, names, pricing, or internal assessments, increasing the sensitivity of undisclosed logging.
Content: No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium

Category: Not specified by scanner
Confidence: 92% confidence
Finding: The presence of generic config, status, and reporting-style capabilities is unjustified for a competitor-analysis skill and indicates unnecessary functionality. Extra commands increase attack surface and user confusion, especially when paired with persistent local storage, because they enable behavior outside the declared scope.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.