Versioned security assessment

Report ID: SA-9AEF3A21

5/23/2026, 8:50:01 AM

higgsfield-generate security assessment v1

Skill Security Certification Report

Audit History

Audit model: claude Historical report

Skill name: higgsfield-generate
Version: v1
Maintainer: higgsfield-ai
Source: github.com/higgsfield-ai/skills/tree/main/higgsfield-generate/
Coverage: 12 Files scanned · 1,107 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.

All skill files are Markdown documentation (.md) containing CLI command examples. The static scanner produced 849 findings but all are false positives caused by misinterpreting Markdown code formatting as executable code patterns. Backticks in Markdown are inline code syntax, not Ruby shell execution. The curl-to-sh bootstrap pattern at SKILL.md:39 is a standard CLI installation convention from the project's own GitHub repository. No malicious intent, credential exfiltration, or data theft vectors were found. The skill legitimately wraps the official higgsfield CLI for image and video generation.

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

12 Files scanned · 1,107 Lines analyzed

1 item 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.

Source

Binding unavailable
Artifact

Identity incomplete
Audit

Complete
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 11 evidence locations

Filesystem access

May read or write local files.

Observed in 4 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 13 evidence locations

Capability review items (1)

Medium

SKILL.md:39

Pipe to shell bootstrap pattern

TRUE POSITIVE (low severity). SKILL.md:39 contains `curl -fsSL https://raw.githubusercontent.com/higgsfield-ai/cli/main/install.sh | sh`. This is a curl-to-sh bootstrap pattern for installing the higgsfield CLI. While standard in developer tooling, this pattern carries inherent risk as it downloads and executes a remote script. However, the URL points to the skill author's own GitHub repository and this is a documented one-time setup step. Risk is low but worth documenting as a known pattern.

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: claude

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 (7)

Medium

SKILL.md:20-297 references/marketing-ad-references.md:9-99 references/marketing-avatars.md:14-46 references/marketing-brand-kits.md:9-53 references/marketing-dtc-ads.md:5-88 references/marketing-modes.md:3-51 references/marketing-products.md:7-53 references/marketing-setup-items.md:3-63 references/media-inputs.md:7-94 references/model-catalog.md:3-159 references/prompt-engineering.md:16-44 references/troubleshooting.md:5-32

Ruby/shell backtick execution in Markdown files

FALSE POSITIVE. Static scanner flagged backtick usage across all 12 Markdown files as Ruby shell backtick execution. In Markdown files, backticks denote inline code formatting, not shell execution. All flagged lines contain CLI command examples shown in code blocks (e.g., `higgsfield generate create ...`) for user reference. These are documentation that instructs the AI agent, not Ruby scripts. 756 locations flagged, all false positives.

Medium

references/marketing-ad-references.md:37-38 references/marketing-ad-references.md:84-90 references/marketing-avatars.md:24-25 references/marketing-avatars.md:34 references/marketing-brand-kits.md:27 references/marketing-products.md:8 references/marketing-products.md:26-27 references/marketing-setup-items.md:43 SKILL.md:198-199

Shell command substitution in Markdown

FALSE POSITIVE. Static scanner flagged $(...) patterns as shell command substitution. These appear inside Markdown code blocks as documented CLI usage examples (e.g., `UPLOAD_ID=$(higgsfield upload create ...)`). The dollar-paren syntax is shown as a usage pattern for readers to understand, not executed code. No active shell injection risk.

Low

Weak cryptographic algorithm in Markdown

FALSE POSITIVE. Static scanner flagged 20+ lines as 'weak cryptographic algorithm'. The lines include Markdown headings like '# Marketing Studio Modes' (line 1), table formatting characters ('---', '|'), and CLI flag descriptions. No cryptographic operations exist in these documentation files. The scanner likely confused Markdown syntax elements with base64 encoding or cryptographic patterns.

Low

System reconnaissance commands in documentation

FALSE POSITIVE. Static scanner flagged CLI commands like `higgsfield account status`, `higgsfield model list`, and `higgsfield marketing-studio avatars list` as 'system reconnaissance'. These are standard CLI discovery commands for the Higgsfield platform, used to check authentication status and list available resources. This is normal operation for any CLI-based skill, not reconnaissance behavior.

Low

references/marketing-brand-kits.md:27-31 references/marketing-products.md:8 references/marketing-products.md:18 references/marketing-products.md:42-48 SKILL.md:39 SKILL.md:153 SKILL.md:224-228 SKILL.md:259

Hardcoded URLs in documentation examples

FALSE POSITIVE. Static scanner flagged example URLs in documentation. These are demonstration URLs (e.g., `https://drinkolipop.com`, `https://shop.example.com/sneakers`, `https://cdn/screenshot1.png`) shown in Markdown code blocks to illustrate CLI usage. No actual network calls are made by these URLs; they are documentation examples only.

Low

references/marketing-avatars.md:34 references/marketing-setup-items.md:43 SKILL.md:198-199

Temp file creation in code examples

FALSE POSITIVE. Static scanner flagged `mktemp` invocations in code examples as temporary file creation. These appear in documented CLI workflows where temp files are needed to pass JSON arrays (e.g., `AVATARS_JSON=$(mktemp)`). This is proper bash practice for creating temporary files, not a security concern.

Low

references/marketing-products.md:3-5 references/troubleshooting.md:24 SKILL.md:223

Fetch API and network calls in Markdown examples

FALSE POSITIVE. Static scanner flagged `fet` and network-related patterns in reference files. These are documented CLI command examples (e.g., `higgsfield marketing-studio products fetch --url ...`) that the AI agent would execute through the CLI, not raw HTTP requests from the skill code. The skill correctly routes all API calls through the official higgsfield CLI as specified in project conventions.

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