ClawHub
Back to skill

Security audit

gladia-using-cli

Security checks for vulnerabilities and agentic risk

Overview

The skill has a coherent transcription purpose, but its setup asks users to run a mutable remote installer script directly in the shell.

Before installing, do not blindly run the curl-to-sh command. Download the installer first, inspect it, prefer a pinned or package-manager installation method if available, and only upload audio or URLs you are authorized to send to Gladia. Be aware that gladia auth set stores an API key locally in ~/.gladia.

A.I.G

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

SkillSpector

Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Tool MisuseTool Parameter Abuse, Chaining Abuse, Unsafe Defaults
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (5)

Chaining Abuse

High
Category
Tool Misuse
Confidence
99% confidence
Finding

The | sh construct is a classic command-chaining abuse pattern because it turns network input directly into shell commands without inspection. In this skill context, the danger is elevated because the skill is designed to be followed by agents or users in a terminal workflow, making unsafe copy-paste execution more likely.

Content

Scanner excerpt · SKILL.md (reported line 42)May include surrounding context.

Install (macOS/Linux):

bash
curl -fsSL https://github.com/gladiaio/gladia-cli/releases/latest/download/install.sh | sh

Auth — get a key at app.gladia.io/account. Three options:

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The skill explicitly states that local files are auto-uploaded and URLs are sent to Gladia's external API, but it does not include a clear privacy or data-handling warning before encouraging use. This can cause users or downstream agents to transmit sensitive audio or private URLs off-host without informed consent, especially in enterprise or regulated environments.

Content

No source excerpt is available for this finding.

Missing User Warnings

Medium
Category
Not specified by scanner
Confidence
98% confidence
Finding

The install command downloads a remote script and immediately executes it with sh, giving the remote content full code-execution capability on the user's machine. If the upstream resource, transport, or release channel is compromised, this becomes an easy path to arbitrary command execution.

Content

No source excerpt is available for this finding.

Whitespace Padding

Medium
Category
Prompt Injection
Confidence
70% confidence
Finding

Large whitespace padding was detected (a block of blank lines or a long run of spaces). This can push injected instructions below or to the right of the visible area so a human reviewer never sees them while the agent still reads them. Manual review of the hidden content is recommended.

Content

Scanner excerpt · SKILL.md (reported line 121)May include surrounding context.

md
| Feature                                        | CLI                              | If not in CLI                                                                          |
| ---------------------------------------------- | -------------------------------- | -------------------------------------------------------------------------------------- |
| Basic transcription                            | `gladia transcribe`              | —                                                                                      |
| Speaker diarization                            | `--diarize`                      | [gladia-audio-intelligence](../gladia-audio-intelligence/SKILL.md) for advanced config |
| Language / code-switch                         | `--language`, `--code-switching` | SDK for advanced `language_config`                                                     |
| Translation, NER, PII, sentiment, audio-to-LLM | No                               | [gladia-audio-intelligence](../gladia-audio-intelligence/SKILL.md)                     |

External Script Fetching

Low
Category
Supply Chain
Confidence
97% confidence
Finding

The skill instructs fetching an external installer script directly from the internet, which creates trust in mutable remote code at execution time. Even over HTTPS, compromise of the hosting account, repository, release pipeline, or DNS/TLS trust chain could lead to malicious code delivery.

Content

Scanner excerpt · SKILL.md (reported line 42)May include surrounding context.

Install (macOS/Linux):

bash
curl -fsSL https://github.com/gladiaio/gladia-cli/releases/latest/download/install.sh | sh

Auth — get a key at app.gladia.io/account. Three options:

Static analysis

No suspicious patterns detected.