Versioned security assessment

Report ID: SA-6B14FCB9

6/30/2026, 4:14:03 AM

agent-ui security assessment v2

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
agent-ui
Version
v2
Maintainer
inference-sh-9
Source
github.com/inference-sh-9/skills/tree/main/skills/agent-ui/
Coverage
1 Files scanned · 120 Lines analyzed
Policy version
Unavailable

Confirmed finding summary

No confirmed security findings

The completed audit recorded no confirmed security findings. This is not proof that the Skill has no side effects.

Installation context

Historical evidence

This report may not describe the currently installable artifact. Open the current Skill page for install guidance.

Open current Skill page

This report does not block or authorize the manifest or ZIP.

Static command-execution and weak-cryptography alerts are mostly false positives caused by Markdown code fences and ordinary text. The skill still carries medium supply-chain and secret-handling risk because it instructs users to fetch remote component code, install an SDK package, and configure an API key backed proxy.

Report position

Historical report

Open audit history before using this report to install.

Audit attestation

Not attestable

The required immutable binding is incomplete.

Human verification

Not verified

No human verification is recorded for this report.

Coverage

1 Files scanned · 120 Lines analyzed

3 items shown for review

Limitations

This report does not claim runtime or sandbox execution and does not prove the absence of side effects.

Evidence chain

Follow the evidence from source binding to the install contract. Available evidence supports verification; it is not a safety guarantee.

  1. Source

    Binding unavailable

  2. Artifact

    Identity incomplete

  3. Audit

    Complete

  4. Install contract

    Open manifest to verify

    Open manifest

Capabilities observed

Observed means this report recorded supporting evidence. Not recorded does not prove that a capability is absent.

Contains scripts

May execute code included with the Skill.

Not recorded by this audit

Network access

May connect to external services.

Observed in 9 evidence locations

Filesystem access

May read or write local files.

Not recorded by this audit

Env variables

May read values from the process environment.

Observed in 1 evidence location

External commands

May invoke commands or programs outside the Skill.

Observed in 2 evidence locations

Capability review items (3)
Medium
SKILL.md:14-19
Remote Component and Package Installation
The Quick Start asks users to run npx shadcn with a remote registry URL and install @inferencesh/sdk. This is a legitimate setup path, but it creates supply-chain risk because remote code and package install behavior must be trusted before execution.
The commands are directly documented and fetch code from an external registry or package source. I found no evidence that the package is malicious, so the concern is supply-chain exposure rather than confirmed abuse.
Medium
SKILL.md:24-36
API Key Proxy Configuration
The setup creates a Next.js proxy route and stores INFERENCE_API_KEY in .env.local. This is expected for the service, but it handles credentials and forwards application traffic through a generated route.
The API key and proxy route are explicit in the instructions. I found no evidence of exfiltration, but incorrect route exposure or key handling would affect user secrets.
Low
SKILL.md:8-10SKILL.md:113-119
External URLs Are Documentation and Asset References
Several hardcoded URLs point to the product site, an image asset, documentation, and blog pages. These links are external network references, but they are visible documentation references rather than hidden outbound requests.
The URLs are visible Markdown links or image references. I found no evidence that the skill silently sends data to these URLs.

Risk findings

Confirmed security concerns are separated from items that still need review.

No confirmed security findings were recorded for this completed audit.

Expert evidence

Immutable subject identity, scanner metadata, dismissed matches, and source-level evidence.

Artifact subject

Marketplace commit
Unavailable
Content hash
Unavailable
Tree hash
Unavailable
Skill path
Unavailable
Audit payload hash
Unavailable

Analysis metadata

Audit model: codex

Analysis state: Complete

Scope is limited to the recorded files, lines, methods, and evidence. No runtime or sandbox execution is claimed.

Static false positives ignored (2)
Low
SKILL.md:14-20SKILL.md:26-56SKILL.md:71-96SKILL.md:100-109
Markdown Backticks Misclassified as Shell Execution
The static analyzer flagged many Markdown code fences and inline prop names as Ruby or shell backtick execution. SKILL.md is documentation, and these backticks do not execute by themselves.
The flagged syntax appears inside Markdown fences or inline documentation. There is no Ruby file, shell script, eval path, or automatic execution mechanism in the scanned file.
Low
SKILL.md:3SKILL.md:50SKILL.md:60SKILL.md:90
Weak Cryptography Alerts Not Confirmed
Static weak-cryptography alerts at the listed lines do not correspond to cryptographic code. The lines contain metadata, descriptive prose, a model reference, or a Markdown table heading.
I inspected each reported line and found no hash, cipher, signature, random-number, or certificate implementation. These are false positives from text scanning.

Verify and export

JSON report CITATION.cff SARIF 2.1.0 Signed installation manifest

The manifest and lockfile bind install artifacts to cryptographic hashes. This integrity claim is separate from the security assessment.

Audit attestation: not_attestable