Versioned security assessment

Report ID: SA-85B53620

6/30/2026, 12:43:13 AM

firebase-basics security assessment v2

Skill Security Certification Report

Audit History
Audit model: codex Historical report
Skill name
firebase-basics
Version
v2
Maintainer
firebase
Source
github.com/firebase/agent-skills/tree/main/skills/firebase-basics/
Coverage
16 Files scanned · 691 Lines analyzed
Policy version
Unavailable

Highest confirmed finding severity

High

3 confirmed security findings require attention.

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 findings are mostly true positives for documented command execution, network access, filesystem configuration, and Firebase API key placeholders, but most are legitimate for a Firebase setup skill. Human review is still required because the skill contains prompt-control language and instructs agents to install remote packages and modify local AI agent configuration files.

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

16 Files scanned · 691 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 6 evidence locations

Filesystem access

May read or write local files.

Observed in 7 evidence locations

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 208 evidence locations

Risk findings

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

Confirmed security concerns (3)

RISK-001 High
SKILL.md:3references/local-env-setup.md:65
Prompt Injection Style Agent Control
The skill uses broad agent-control language that tells agents to always use the skill for Firebase and not proceed with Firebase tasks until setup is complete. This does not show confirmed malicious intent, but it can override normal task prioritization and requires marketplace review.
The cited lines contain explicit always-use and do-not-proceed agent directives. They do not target the evaluator or security review, so confidence is high but not critical.
RISK-002 Medium
SKILL.md:10-15references/local-env-setup.md:45-51references/refresh-other-agents.md:41-48
Remote Package Execution Through CLI Guidance
The skill repeatedly instructs agents to run npx commands that fetch the latest Firebase tools or skills packages. This is expected for Firebase setup, but it creates supply-chain and unattended execution risk.
The commands are explicit and semantically intended to execute remote package tooling. The behavior is legitimate for the skill, but it remains a real operational risk.
RISK-003 Medium
references/setup-cursor.md:24-28references/setup-antigravity.md:24-28references/setup-other_agents.md:28-30
Local Agent Configuration File Modification
The setup guides instruct agents to locate, create, and merge MCP server configuration into local or home-directory agent settings. This is legitimate setup behavior, but incorrect edits could alter trusted tool configuration.
The files clearly direct agents to modify MCP configuration files, including hidden home-directory paths. The instructions also say to preserve existing servers, reducing but not eliminating risk.

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 (3)
Low
references/local-env-setup.md:13references/setup-gemini_cli.md:15references/refresh-other-agents.md:7
Hardcoded Documentation URLs
The network findings are links to official or relevant setup documentation and repository locations. They do not show hidden callbacks or exfiltration endpoints.
The URLs are plainly shown in Markdown setup instructions. No evidence found that these links transmit local data or credentials.
Low
references/web_setup.md:36-45
Firebase Web API Key Placeholder
The env_access finding points to an example Firebase web configuration placeholder. Firebase web API keys identify a project but are not service account secrets by themselves.
The value is a placeholder inside a standard Firebase client initialization example. No evidence found of real credentials or exfiltration.
Low
SKILL.md:3references/setup-claude_code.md:13-14references/setup-gemini_cli.md:12
Static Weak-Crypto Detections Are Text Matches
The weak cryptographic algorithm findings appear to be false positives triggered by ordinary words and command text in setup documentation. No evidence found of implemented cryptographic code.
The cited content is Markdown setup guidance, not executable crypto implementation. The static pattern likely matched substrings such as skill names or command arguments.

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