Back to skill

Security audit

Aha Codex Omx

Security checks for vulnerabilities and agentic risk

Overview

This skill is a disclosed orchestration guide for using Codex with OMX, with no hidden code or automatic persistence found.

Before installing or upgrading, prefer a reviewed release or commit instead of blindly running the latest git pull or npx command. If you use this skill, expect it to encourage OMX capability discovery and bounded delegation, while still requiring confirmation for high-cost, persistent, worktree, or autonomous activity.

Vulnerability Patterns
  • Insecure DependenciesIntroduces malicious components through unsafe dependency sources
  • 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
Findings (1)

T08 · Insecure Dependencies

Warning
Location
SKILL.md:76
Finding

Unpinned Installer and Mutable Repository Used for Skill Upgrades

Content
View full analysis

Vulnerability Details

File Location: SKILL.md, line 76
Vulnerability Type: T08: Insecure Dependencies
Risk Level: Medium

Vulnerable Code

markdown
- **Upgrade**: Run `git pull` in the aha-orch repo, or re-run `npx skills add its-how/aha-orch` to get the latest version.

Technical Analysis

The documented upgrade process retrieves components from mutable upstream sources without pinning the installer version, repository release, or commit SHA. It also does not require checksum or signature verification.

Running npx skills add its-how/aha-orch may obtain and execute the currently published version of the relevant npm tooling, while git pull incorporates whatever content is present on the tracked upstream branch. Consequently, the effective upgrade behavior and installed Skill content can change after this version has been audited.

This creates a supply-chain trust dependency on the npm publisher, package distribution infrastructure, GitHub repository, maintainer accounts, and upstream branch. A compromise of any relevant upstream publishing channel could cause a subsequent upgrade to install unreviewed or malicious content.

Attack Path

  1. An attacker compromises a relevant npm publisher account, package distribution channel, GitHub maintainer account, or the upstream repository.
  2. The attacker publishes malicious installer behavior or modifies the mutable repository branch with malicious Skill instructions or supporting content.
  3. A user follows the documented upgrade procedure and runs either the unpinned npx command or git pull.
  4. The altered upstream content is downloaded without version pinning or integrity verification.
  5. In the npx case, malicious package behavior may execute during installation under the invoking user's permissions. Malicious Skill content installed from either route may subsequently influence agent sessions when loaded.

Impact Assessment

Ex ...[truncated 682 chars]

Remediation
View remediation

Remediation Suggestions

  1. Pin the npm installer to a specifically reviewed version instead of allowing npx to resolve the latest release.
  2. Pin Skill installation and upgrades to a signed release tag or immutable commit SHA.
  3. Publish cryptographic checksums or signatures for release artifacts and verify them before installation.
  4. Replace unrestricted git pull guidance with a process that fetches a specific release and displays the proposed changes for review before activation.
  5. Use npm lockfiles or equivalent dependency-locking controls where applicable, and require integrity metadata for resolved packages.
  6. Run installation with the least privileges necessary in a restricted environment that does not expose credentials or sensitive workspace data.
  7. Require manual review of changes to Skill instructions, scripts, permissions, and external dependencies before enabling an upgraded version.
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Rp1

Medium
Category
MCP Rug Pull
Confidence
90% confidence
Finding

The skill instructs users to run npx skills add its-how/aha-orch without pinning a specific version or commit. This can cause users to fetch and execute whatever package version is current at install time, creating a supply-chain risk if a malicious or compromised release is published later.

Content

No source excerpt is available for this finding.

Natural-Language Policy Violations

Low
Category
Not specified by scanner
Confidence
78% confidence
Finding

Lines L239-L245 state that verbosity:low governs every transparency statement and impose strict brevity limits. This is a natural-language policy constraint on communication style without indicating user choice or opt-in, which matches the policy-violation category for forced language/locale-style behavior.

Content

No source excerpt is available for this finding.