Security audit

frontend-design-deslop

Security checks for vulnerabilities and agentic risk

Overview

This is an opinionated frontend design helper with disclosed local project edits and no evidence of hidden, destructive, or credential-seeking behavior.

Use this if you want a strong, opinionated frontend design workflow. Expect it to read and edit project UI files and create or update DESIGN.md. Review its design choices before committing them, and do not ship the optional unpinned unpkg Open Props import unless you pin or vendor the dependency and accept the external runtime CSS dependency.

A.I.G

Vulnerability Patterns

Insecure DependenciesIntroduces malicious components through unsafe dependency sources
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

Findings (1)

T08 · Insecure Dependencies

Warning

Location: references/adapters.md:78
Finding: Unpinned Runtime CSS Dependency from a Third-Party CDN
Content: View full analysis
Remediation: View remediation
` with a verified Subresource Integrity hash and `crossorigin="anonymous"`. - CSS `@import` does not provide an equivalent SRI attribute, so it should not be recommended for security-sensitive remote dependencies. 4. **Restrict browser network access** - Configure a Content Security Policy with a narrow `style-src`. - Restrict `img-src`, `font-src`, and related directives to prevent imported CSS from freely loading attacker-controlled secondary resources. 5. **Use local first-party tokens by default** - Make `references/token-core.css` the default and sufficient source of design tokens. - Remove the remote import recommendation unless the user explicitly requests Open Props and accepts the external dependency. 6. **Add dependency governance** - Track the vendored or pinned release in dependency review tooling. - Review upstream changes before upgrades. - Test visual integrity and application behavior after each approved update. ]]>

SkillSpector

Vulnerability Patterns

Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (3)

Vague Triggers

Medium

Category: Not specified by scanner
Confidence: 95% confidence
Finding: The invocation guidance covers extremely broad situations such as 'whenever building or styling any web frontend, app, dashboard, landing page, deck, or artifact,' which overlaps with many routine development tasks. It does not clearly define boundaries or exclusions for when the skill should not activate, increasing the risk of unintended invocation.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 92% confidence
Finding: Line L100 specifically singles out French copy and instructs the agent to defer to named French-language humanization skills. This imposes a locale-specific workflow preference in natural language without indicating user choice, opt-in, or a documented justification for the language constraint.
Content: No source excerpt is available for this finding.

Natural-Language Policy Violations

Low

Category: Not specified by scanner
Confidence: 88% confidence
Finding: Line L07 states "never Inter/Roboto/system as primary," which is a natural-language instruction enforcing a specific typography choice. This is a policy-like constraint without any indication of user opt-in or a documented region/compliance justification, so it fits the language/style policy concern defined for natural-language violations.
Content: No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.