ClawHub
Back to skill

Security audit

present

Security checks across malware telemetry and agentic risk

Overview

The skill is mostly purpose-aligned for creating and publishing narrated presentations, but it deserves review because it recommends running unpinned remote installer scripts and can upload user content using Bisque credentials.

Review the installer source and prefer a verified or package-managed installation path before running the bisque-voice install/update commands. Install only if you are comfortable giving this skill Bisque account access and allowing it to upload the deck HTML, assets, narration audio, and any context.md content at the selected presentation visibility.

SkillSpector

Vulnerability Patterns
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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 (5)

Vague Triggers

Medium
Confidence
93% confidence
Finding
The skill advertises activation on very broad phrases like 'make a presentation', 'slides', 'video', 'deck', 'slideshow', and 'talk', which creates a real risk of over-triggering on loosely related user requests. Because this skill can lead to publishing content and running local commands, accidental invocation expands the chance of unintended side effects beyond simple misclassification.

External Script Fetching

High
Category
Supply Chain
Content
"Install it, then re-run:\n" +
      (process.platform === "win32"
        ? "  irm https://download.bisque.today/bisque-voice/install.ps1 | iex"
        : "  curl -fsSL https://download.bisque.today/bisque-voice/install.sh | sh"),
  );
}
Confidence
92% confidence
Finding
The script instructs users to install software by piping a remotely fetched shell script directly into `sh` (or `iex` on Windows). If the download host, transport, upstream artifact, or DNS/TLS trust path is compromised, users can execute arbitrary code immediately on their machine.

External Script Fetching

High
Category
Supply Chain
Content
say(
      "note: this bisque-voice has no aligner support — narration sync may " +
        "drift on Studio engines. Update it: " +
        "curl -fsSL https://download.bisque.today/bisque-voice/install.sh | sh",
    );
    return;
  }
Confidence
92% confidence
Finding
This message again recommends updating via a remote script piped straight to the shell. Even though the script does not auto-execute it, this is still dangerous in a developer tool because users are likely to copy-paste the exact remediation text during troubleshooting.

External Script Fetching

High
Category
Supply Chain
Content
say(
      process.platform === "win32"
        ? "  install: irm https://download.bisque.today/bisque-voice/install.ps1 | iex"
        : "  install: curl -fsSL https://download.bisque.today/bisque-voice/install.sh | sh",
    );
  }
  const auth = resolveAuth({
Confidence
93% confidence
Finding
The doctor command prints installation instructions that fetch and execute a remote script in one step. In this skill context, the advice is especially risky because the tool handles credentials and local files, so compromise of the install path can expose API keys and user content.

External Script Fetching

High
Category
Supply Chain
Content
"update it first:\n" +
            (process.platform === "win32"
              ? "  irm https://download.bisque.today/bisque-voice/install.ps1 | iex"
              : "  curl -fsSL https://download.bisque.today/bisque-voice/install.sh | sh"),
        );
      }
      fail(
Confidence
92% confidence
Finding
This update advice repeats the same unsafe pattern of executing network-fetched code directly. Because it appears in an error path, users may be under pressure to fix the problem quickly and be more likely to run the command without scrutiny.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

Detected: suspicious.dangerous_exec, suspicious.env_credential_access, suspicious.exposed_secret_literal

Shell command execution detected (child_process).

Critical
Code
suspicious.dangerous_exec
Location
scripts/present.mjs:469

Environment variable access combined with network send.

Critical
Code
suspicious.env_credential_access
Location
scripts/present.mjs:43

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
scripts/present.mjs:899