The skill facilitates arbitrary code execution by piping unverified remote scripts directly into a shell and lacks necessary security declarations to constrain its network and system access.
npx skills add https://github.com/EximAgent/cliThe skill documentation explicitly instructs the agent to download and execute remote code directly into a shell interpreter via `curl | sh` or `irm | iex`. This bypasses security controls and allows for arbitrary code execution on the host environment.
Install instructions: 'curl -fsSL https://cli.eximagent.ai/install | sh' and 'irm https://cli.eximagent.ai/install.ps1 | iex'.
Remote code download and execution detected
curl -fsSL https://cli.eximagent.ai/install | sh
A remote resource is downloaded with curl/wget and piped directly into a shell interpreter (optionally via sudo). The downloaded script runs with no integrity check (no checksum/signature), so a compromised server, MITM, or hijacked URL leads to arbitrary code execution. Download to a file, verify it (e.g. sha256sum -c against a pinned digest, or a signature), then execute it.
curl -fsSL https://cli.eximagent.ai/install | sh
The installation instructions use `curl | sh` and `irm | iex`, which are dangerous patterns that execute remote code with full user privileges without integrity verification.
curl -fsSL https://cli.eximagent.ai/install | sh
NER model detected location in skill content (confidence: 0.62)
U*
NER model detected location in skill content (confidence: 1.00)
G******
NER model detected organization in skill content (confidence: 0.58)
m****
NER model detected organization in skill content (confidence: 0.71)
O***
The documentation explicitly instructs users to bypass sandbox network restrictions to allow the CLI to function, potentially exposing the host environment to unauthorized egress. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
codex --dangerously-bypass-approvals-and-sandbox
NER model detected organization in skill content (confidence: 0.56)
N**
The skill includes commands like `crawl run` and `enrich company` which fetch and process external website content. If the agent follows instructions found within the crawled content, it is vulnerable to indirect prompt injection. [ensemble: confirmed by 3/3 passes; severity set to the agreed median (ADR-0067).]
eximagent crawl --inputs -
Skill does not specify a license field. Specifying a license helps users understand usage terms.
[](https://mondoo.com/ai-agent-security/skills/github/EximAgent/cli/eximagent)<a href="https://mondoo.com/ai-agent-security/skills/github/EximAgent/cli/eximagent"><img src="https://mondoo.com/ai-agent-security/api/badge/github/EximAgent/cli/eximagent.svg" alt="Mondoo Skill Check" /></a>https://mondoo.com/ai-agent-security/api/badge/github/EximAgent/cli/eximagent.svgSkills can read files, run commands, and access credentials. Mondoo helps organizations manage the security risks of AI agent skills across their entire fleet.