Back to skill

Security audit

Slickdeals

Security checks for vulnerabilities and agentic risk

Overview

This is a shopping deal-search helper with expected web access and some quality/scope caveats, but no evidence of hidden, destructive, persistent, credential-stealing, or exfiltrating behavior.

Install only if you are comfortable with shopping queries being sent to Slickdeals and, in some cases, retailer websites for availability or price checks. Treat retailer pages as untrusted web content and verify prices before buying; also note that the included category option appears not to work as advertised.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Findings (7)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
90% confidence
Finding

The skill requires network access via ollama_web_fetch but does not declare any explicit tool scope such as permissions or allowed-tools. This weakens least-privilege controls and makes the skill's actual runtime capabilities less transparent to the host, increasing the chance of unintended or overly broad web access.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
93% confidence
Finding

The manifest description says to use the skill when the user asks for deals, discounts, price comparisons, or cheap products, which is a very broad class of everyday shopping requests. It does not clearly bound when this skill should be invoked instead of more general shopping or comparison behavior, increasing the risk of unintended activation.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill extends beyond Slickdeals by instructing the agent to check retailer pages for availability and pricing. This broadens the trust boundary and causes the agent to fetch and act on content from arbitrary third-party sites, which can increase exposure to prompt injection, tracking, or off-scope data retrieval.

Content

No source excerpt is available for this finding.

Description-Behavior Mismatch

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The fallback guidance explicitly tells the agent to supplement Slickdeals results with current retail pricing from Amazon, Best Buy, and other sites when Slickdeals has no active deals. This is a clear scope expansion from a single-site search skill into general web shopping behavior, increasing the chance of unintended browsing and untrusted-instruction ingestion from external pages.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The module defines a Deal dataclass and a SlickdealsParser with docstrings implying structured parsing of Slickdeals search results, and search_deals is documented as returning a List[Deal]. In practice, search_deals ignores the parser and returns a list of raw text strings extracted with regexes, so the code documentation and type intent actively misrepresent what the function actually produces.

Content

No source excerpt is available for this finding.

Missing User Warnings

Low
Category
Not specified by scanner
Confidence
88% confidence
Finding

This code sends the user-provided query to an external website via an HTTP request. Although network access is central to the skill’s purpose, the file lacks any explicit user-facing disclosure such as a print/log message or comment stating that the query will be sent to Slickdeals.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
98% confidence
Finding

The command-line interface documents a '--category' option as a category filter, which tells users the skill can constrain results by category. However, args.category is never used when building the request or filtering results, so the inline interface documentation contradicts the actual behavior.

Content

No source excerpt is available for this finding.