Versioned security assessment

Report ID: SA-803C9C36

7/9/2026, 5:41:47 AM

nano-banana-2 security assessment v7

Skill Security Certification Report

Audit History
Audit model: claude Historical report
Skill name
nano-banana-2
Version
v7
Maintainer
agentspace-so
Source
github.com/agentspace-so/runcomfy-agent-skills/tree/main/nano-banana-2/
Coverage
1 Files scanned · 207 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.

The skill is primarily Markdown guidance for running Google Nano Banana 2 through the RunComfy CLI. Confirmed risks are limited to disclosed external CLI execution and outbound RunComfy API access; most static hits are false positives from code fences, tables, links, or security documentation.

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 · 207 Lines analyzed

10 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 4 evidence locations

Filesystem access

May read or write local files.

Observed in 2 evidence locations

Env variables

May read values from the process environment.

Not recorded by this audit

External commands

May invoke commands or programs outside the Skill.

Observed in 40 evidence locations

Capability review items (10)
Medium
SKILL.md:13
Ruby/shell backtick execution
`runcomfy run google/nano-banana-2/text-to-image` through the local
The skill explicitly states that it calls the local RunComfy CLI. This is an expected capability, but it is still external command execution that can submit remote jobs.
Medium
SKILL.md:50-51
Ruby/shell backtick execution
1. **RunComfy CLI** — `npm i -g @runcomfy/cli`
This line instructs users to install the RunComfy CLI with npm. Installing and running a third-party CLI is a real external-command risk even though it is disclosed.
Medium
SKILL.md:51-52
Ruby/shell backtick execution
2. **RunComfy account** — `runcomfy login` opens a browser device-code flow.
This line instructs users to run runcomfy login. It is legitimate setup, but it executes a local CLI and initiates an authentication flow.
Medium
SKILL.md:52
Ruby/shell backtick execution
3. **CI / containers** — set `RUNCOMFY_TOKEN=<token>` instead of `runcomfy login`.
This line documents RUNCOMFY_TOKEN for CI use. It is a legitimate credential path but confirms the skill depends on external CLI authentication.
Medium
SKILL.md:76-80
Ruby/shell backtick execution
```bash
The surrounding block is an executable runcomfy command template. Arguments are shown as structured JSON, but the skill still runs an external command.
Medium
SKILL.md:84-93
Ruby/shell backtick execution
```bash
The surrounding block is an executable runcomfy batch-generation command. It is expected behavior, but it invokes a third-party CLI.
Medium
SKILL.md:97-106
Ruby/shell backtick execution
```bash
The surrounding block is an executable runcomfy command for final image generation. It is documented and structured, but still executes a CLI.
Medium
SKILL.md:110-117
Ruby/shell backtick execution
```bash
The surrounding block is an executable runcomfy command with web grounding enabled. This is expected, but it invokes a CLI and may trigger network access.
Medium
SKILL.md:198
Ruby/shell backtick execution
The skill invokes `runcomfy run google/nano-banana-2/text-to-image` with a JSON body matching the sc
This line describes the actual runtime behavior: invoking runcomfy run and submitting the request. That confirms external command execution and remote API use.
Low
SKILL.md:198
Hardcoded URL
The skill invokes `runcomfy run google/nano-banana-2/text-to-image` with a JSON body matching the sc
This line documents the RunComfy Model API endpoint used for request submission. It is disclosed and expected, but it is real outbound network access.

Risk findings

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

No confirmed security findings were recorded for this completed audit.

Remediation

Suggested fixes recorded by this audit. Applying them is the maintainer’s responsibility.

  1. FIX-001
    Medium
    External CLI execution is central to the skill.
    Keep the command arguments structured as JSON, require user confirmation before running generation jobs, and avoid interpolating untrusted text into shell syntax.
  2. FIX-002
    Low
    RunComfy credentials are required for authenticated use.
    Recommend RUNCOMFY_TOKEN for CI and remind users not to paste tokens into prompts, logs, or generated examples.
  3. FIX-003
    Low
    Generated outputs are downloaded from remote RunComfy domains.
    Keep the documented output-domain whitelist and file-size cap, and store generated files only in user-selected output directories.

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: claude

Analysis state: Complete

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

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