Back to skill

Security audit

SlideClaw

Security checks for vulnerabilities and agentic risk

Overview

The skill is coherent for making SlideClaw/Marp slide decks, but it can run an unpinned npm package and make persistent or out-of-workspace file changes without clear opt-in.

Review this skill before installing. It appears intended for legitimate SlideClaw/Marp deck creation, but users should avoid the `npx -y slideclaw@latest` fallback unless they trust the current npm package, prefer a pinned or locally installed version, use `--no-copy-to-downloads` or `SLIDECLAW_NO_DOWNLOADS=1` if they do not want files copied outside the workspace, and approve any `profile/user.md` updates explicitly.

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

Error
Location
SKILL.md:24
Finding

Automatic Execution of an Unpinned npm Package

Content
View full analysis
Remediation
View remediation
``` 2. Verify the selected release and its transitive dependencies through a committed lockfile and npm integrity hashes. 3. Require explicit user approval before downloading or executing software that is not already installed. 4. Prefer a trusted, locally installed binary whose version is checked against an allowlist. 5. Run package installation and rendering in a sandbox with: - Restricted filesystem access. - Minimal environment variables. - Limited network access. - No elevated privileges. 6. Establish a controlled update process in which new package versions are reviewed and tested before changing the pinned version. 7. Consider using a trusted internal registry or verified package mirror to reduce exposure to public-registry compromise. ]]>
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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 (6)

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill's fallback behavior directs use of npx -y slideclaw@latest, meaning the agent may retrieve and execute remote package code on demand. This is dangerous because the package code is outside the reviewed skill content and can change over time, giving an attacker a supply-chain path to arbitrary command execution.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
95% confidence
Finding

The skill instructs the agent to execute npx -y slideclaw without a pinned version, which can fetch and run whatever package version is current at execution time. That creates a supply-chain risk: a compromised upstream package, malicious new release, or typo/package takeover could lead to arbitrary code execution in the user's environment.

Content

No source excerpt is available for this finding.

Rp1

Medium
Category
MCP Rug Pull
Confidence
99% confidence
Finding

npx -y slideclaw@latest explicitly authorizes downloading and executing the newest remote package release at runtime. In a skill that is meant to run commands autonomously, this is especially risky because it turns normal use into dynamic remote code execution with no version review or stability guarantees.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The skill instructs the agent to create projects and edit files such as requirements.md and slides.md as part of the workflow, but does not clearly require a user-facing warning that these writes will occur. While file creation is expected in a slide-production skill, lack of transparency about modifications can still surprise users and reduces informed consent, especially when combined with later durable profile updates.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The skill describes automatic copying of rendered PDFs into the user's Downloads directory and even tells the agent to prefer reporting that path, but does not require warning or consent for this extra filesystem side effect. Writing to a user-global Downloads folder extends beyond the project workspace and may expose sensitive deck content or create unexpected artifacts in a commonly synced/shared location.

Content

No source excerpt is available for this finding.

Context-Inappropriate Capability

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The skill directs the agent to update profile/user.md with durable preferences, which is unrelated to the narrow task of generating/rendering slides and creates persistent state changes beyond the immediate request. This broadens the skill's authority and can enable unwanted accumulation of user data or prompt-influencing state that persists into future interactions.

Content

No source excerpt is available for this finding.