AfrexAI Pricing Optimizer

• Outcome: Pass
• Latest audit: Sep 10, 2026 · 6:23 PM UTC
• Version: 1.0.0

Overview

This skill is a plain pricing-advice prompt with no hidden execution, persistence, or sensitive access, though it links to optional external resources.

Review the optional external links and only install the related lead-scoring skill if you trust its publisher and current contents. The pricing optimizer itself is a markdown-only advisory skill and does not request unusual authority.

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:75

Finding

Unpinned Third-Party Skill Installation Creates a Supply-Chain Risk

Content

View full analysis

Vulnerability Details

File Location: SKILL.md, line 75
Vulnerability Type: T08: Insecure Dependencies
Risk Level: Medium

Complete Code Snippet:

markdown

- Lead scoring: `clawhub install afrexai-lead-scorer`

Technical Analysis

The Skill recommends installing the third-party afrexai-lead-scorer package by a mutable package name. The command does not specify an immutable version, commit digest, checksum, trusted provenance, or review requirement.

Consequently, the installed content may differ from the content originally intended or reviewed. If the package is compromised, transferred to a malicious maintainer, or updated with unsafe instructions or executable components, users or agents following this recommendation could install attacker-controlled content. The external package was not included in the audited project, so its behavior cannot be verified by this audit.

Attack Path

1. An agent or user follows the related-tool recommendation in SKILL.md.
2. ClawHub resolves the mutable package identifier afrexai-lead-scorer.
3. A compromised or malicious package version is downloaded and installed.
4. The installed Skill supplies attacker-controlled instructions or executable behavior.
5. When invoked, that content operates using the tools and permissions available to the consuming agent or ClawHub environment.

This path depends on the third-party package or its distribution channel being compromised; the audited file itself does not demonstrate malicious package behavior.

Impact Assessment

The potential impact is bounded by the permissions granted to the installed Skill and its host agent. A malicious dependency could influence agent behavior, access data exposed to the agent, invoke available tools, or execute operations permitted by the runtime. No evidence in the audited project establishes privilege escalation, persistence, credential theft, data exfiltration, or malici ...[truncated 37 chars]

Remediation

View remediation

Remediation Suggestions

1. Remove the installation recommendation if the external Skill is not essential.
2. Pin the dependency to an immutable, previously reviewed version or content digest rather than a mutable package name.
3. Document the authoritative publisher and trusted distribution source.
4. Verify package integrity with a cryptographic checksum or signature before installation.
5. Perform a separate security review of the exact pinned package contents, including scripts, instructions, dependencies, and requested permissions.
6. Require explicit user approval before installation or activation.
7. Run the installed Skill with least privilege in a sandbox that restricts filesystem, network, credential, and tool access.
8. Monitor pinned dependency versions for compromise and repeat the security review before upgrading.

SkillSpector

Vulnerability Patterns

• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• 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

Static analysis

No suspicious patterns detected.