Acp Loop

• Outcome: Review
• Latest audit: Sep 11, 2026 · 5:12 PM UTC
• Version: 0.1.0

Overview

The skill does what it says, but it asks users to install a mutable global npm CLI and can run recurring agent prompts indefinitely without strong safety boundaries.

Review the npm package and version before installing, avoid running the install with elevated privileges, and use bounded schedules such as `--max` or `--timeout` for any prompt that could modify files, spend money, contact services, or publish data. Prefer read-only or dry-run prompts unless you are comfortable with repeated autonomous execution.

A.I.G

Vulnerability Patterns

• Insecure Dependencies: Introduces malicious components through unsafe dependency sources
• Skill Instruction Hijacking: Alters the agent's session goals or safety constraints when the skill loads
• Agent Memory Poisoning: Writes attacker-controlled rules into memory that affect later sessions
• Remote Payload Retrieval and Execution: Fetches external code whose behavior can change after review
• Embedded Malicious Code: Ships malicious scripts inside the skill and executes them locally

Findings (1)

T08 · Insecure Dependencies

Warning

• Location: SKILL.md:20

Finding

Unpinned Global Installation of a Third-Party npm Package

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, lines 20–24
Vulnerability Type: Unpinned third-party dependency installed globally
Risk Level: Medium

Vulnerable Code

markdown

## Installation

```bash
npm install -g acp-loop

text

### Technical Analysis

The installation instructions retrieve and globally install the latest version of the third-party `acp-loop` package from the npm registry. The package version and artifact integrity are not pinned, so the code installed by users may differ from the version that was available when this Skill was reviewed.

npm packages can define lifecycle scripts that execute during installation with the permissions of the invoking user. Global installation also places an executable in the user's global command search path. Consequently, compromise of the npm publisher account, package, or distribution process could cause users following these instructions to execute attacker-controlled code or install a malicious replacement CLI.

No evidence was found that the referenced package is currently malicious. The risk arises from the unpinned and mutable supply-chain dependency.

### Attack Path

1. An attacker compromises the `acp-loop` npm package, its publisher account, or its release process.
2. The attacker publishes a malicious release under the package's existing name and latest distribution tag.
3. A user follows the Skill's documented `npm install -g acp-loop` command.
4. npm downloads the attacker-controlled release without enforcing a reviewed version or expected integrity value.
5. Malicious lifecycle scripts may execute during installation, or a malicious `acp-loop` executable may be installed globally.
6. The malicious code executes with the invoking user's privileges when installation occurs or when the installed CLI is subsequently used.

### Impact Assessment

Successful exploitation could permit arbitrary code execution with the privileges of 
...[truncated 357 chars]

Remediation

View remediation

Remediation Suggestions

• Pin the dependency to an exact version that has been reviewed, for example npm install -g acp-loop@<reviewed-version>, rather than relying on the mutable latest release.
• Record and verify the expected package provenance and integrity before installation. Where supported, use npm provenance attestations and verify the publisher and release metadata.
• Prefer a project-local installation with a lockfile over global installation, limiting changes to the user's global executable environment.
• Review package contents and lifecycle scripts before recommending a release.
• If lifecycle scripts are not required, install with --ignore-scripts and document any resulting functionality limitations.
• Advise users not to execute the installation command with sudo, root privileges, or another privileged account.
• Establish a process to periodically reassess the pinned version and deliberately update it only after security review.

SkillSpector

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code
• Excessive Agency: Unrestricted Tool Access, Autonomous Decision Making, Scope Creep

Findings (1)

Missing User Warnings

Medium

• Category: Not specified by scanner
• Confidence: 95% confidence

Finding

This skill explicitly enables unattended recurring execution of prompts via interval and cron scheduling, but the documentation does not warn users that each scheduled run may invoke an agent without fresh confirmation. In context, that can lead to repeated autonomous actions such as cleanup, reporting, status checks, or code review workflows that may affect user data, systems, costs, or external services if the prompt is unsafe or misconfigured.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.