Back to skill

Security audit

Svg2pptskill

Security checks for vulnerabilities and agentic risk

Overview

This skill does what it says: it sends user-provided SVG/map/character input and a user-provided Handbooks API key to Handbooks to generate an editable PowerPoint file.

Install only if you are comfortable sending the SVG content, SVG URLs, map/character request, and Handbooks API key to handbooks.cn. Avoid using sensitive SVGs, and prefer the web UI or stdin/secret-handling patterns over placing API keys directly in shell commands.

Vulnerability Patterns
  • Insecure Skill Coding PracticesFinds exploitable flaws such as hardcoded secrets or command injection
  • 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)

T09 · Insecure Skill Coding Practices

Warning
Location
scripts/convert.cjs:208
Finding

API Key Exposed Through Command-Line Arguments

Content
View full analysis
" })); process.exit(1); } const key = argv[0]; ``` The project documentation actively recommends placing the API key directly in the command: ```bash node scripts/convert.cjs SVG2PPTSKILL-xxx map:安徽 node scripts/convert.cjs SVG2PPTSKILL-xxx char:中 node scripts/convert.cjs SVG2PPTSKILL-xxx map:Anhui ``` ### Technical Analysis The conversion script receives the API key as its first command-line argument and retrieves it through `process.argv`. Secrets passed this way can be exposed through shell history, process inspection facilities, command telemetry, terminal session recording, diagnostic reports, or logs generated by wrapper and orchestration tools. The SVG value is also normally supplied as a command-line argument. Consequently, confidential SVG content may be exposed through the same channels. HTTPS protects the subsequent network request in transit, but it does not prevent local disclosure before the request is sent. Exploitation requires access to a relevant local observation channel, such as the victim's command history, process metadata, terminal logs, or automation telemetry. No unrelated remote exfiltration endpoint was identified in the audited code. ### Attack Path 1. A user follows the documented instructions and invokes the script with an API key in the command line. 2. The shell records the complete command in history, or a local monitoring, telemetry, wrapper, or process-inspection facility capt ...[truncated 1097 chars]
Remediation
View remediation
...' | SVG2PPT_KEY="$SVG2PPT_KEY" node scripts/convert.cjs unset SVG2PPT_KEY ``` For stronger protection, the script should read the key from a non-echoing prompt or protected file descriptor rather than exposing it in either command arguments or a broadly inspectable environment. ]]>
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Undeclared Tool Scope

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding

The skill documents active network use and direct invocation of an external conversion API, but it does not declare any explicit tool scope such as allowed network permissions. That creates a governance and least-privilege gap: an agent may be able to transmit user content and secrets off-platform without clear permission boundaries or reviewable policy metadata.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
97% confidence
Finding

The skill instructs sending user-provided SVG data, URLs, and an API key to a third-party service, but it does not provide an explicit privacy notice or data-handling warning. This is risky because SVG content, URLs, or embedded metadata may contain sensitive information, and the user may not realize their data and credential are being transmitted to an external operator.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.