Versioned security assessment

Report ID: SA-AA5CAC55

6/28/2026, 5:30:32 AM

detection-sigma security assessment v6

Skill Security Certification Report

Audit History

Audit model: codex Historical report

Skill name: detection-sigma
Version: v6
Maintainer: AgentSecOps
Source: github.com/AgentSecOps/SecOpsAgentKit/tree/main/skills/incident-response/detection-sigma
Coverage: 13 Files scanned · 2,873 Lines analyzed
Policy version: Unavailable

Highest confirmed finding severity

Medium

1 confirmed security finding requires 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 analysis reported many critical and high indicators, but most are false positives caused by Sigma detection examples, ATT&CK terminology, markdown command snippets, and reference URLs. One real documentation risk remains: a Splunk deployment example uses placeholder credentials and disables TLS verification. No confirmed malicious intent or prompt injection attempt was found, so publication is acceptable with a security warning.

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

13 Files scanned · 2,873 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 38 evidence locations

Filesystem access

May read or write local files.

Not recorded by this audit

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 469 evidence locations

Risk findings

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

Confirmed security concerns (1)

RISK-001 Medium

references/backend-support.md:265-277

Unsafe Deployment Example Uses Placeholder Credentials and Disabled TLS Verification

The Splunk deployment example defines placeholder admin credentials and sends a REST request with TLS verification disabled. This appears to be documentation, not malicious code, but users could copy it into production automation.

The cited lines clearly show a deployment URL, placeholder basic authentication, and verify=False. The risk is copy-paste misuse rather than confirmed malicious execution.

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 (4)

Low

SKILL.md:19-21 references/log-source-guide.md:140 references/backend-support.md:265-277

Static Network Indicators Are Documentation and References

Hardcoded URLs and the .onion example appear in references, ATT&CK links, backend documentation, and Sigma detection examples. I did not find evidence that the skill initiates network traffic by itself.

The network strings are visibly documentation, resource links, or detection criteria. The only actionable network code is the deployment example already captured as a medium issue.

Low

SKILL.md:42-44 SKILL.md:48-54 references/backend-support.md:7-14

External Command Alerts Are Mostly Markdown Examples

PowerShell, cmd.exe, Python, and shell snippets are used to explain Sigma rule patterns, installation commands, and conversion workflows. They are not hidden execution paths inside the skill package.

The cited examples are fenced documentation blocks and detection examples. Legitimate use still requires users to run tools in their own environment.

Low

assets/rule-templates/credential-access.yml:44-72 references/mitre-attack-mapping.md:312-321 references/field-modifiers.md:155-164

Credential, Ransomware, C2, and Obfuscation Keywords Are Defensive Detection Content

The high-severity keyword alerts map to Sigma rules and ATT&CK reference content for detecting credential access, encoded commands, onion domains, command and control, and ransomware file changes.

The surrounding context is explicitly labeled as Sigma detection patterns and modifier documentation. I did not find instructions to steal credentials, deploy ransomware, or operate C2 infrastructure.

Low

assets/compliance-rules/pci-dss-monitoring.yml:28 assets/rule-templates/persistence.yml:22

No Prompt Injection Attempt Found in Reviewed Suspicious Text

Search terms related to prompt injection matched normal compliance and ATT&CK wording, such as administrator activity and system updates. No evidence found for instructions to override the evaluator or skip security review.

Targeted review of suspicious phrases found benign security-domain language. This does not prove absence across all text, but no prompt-injection evidence was identified.

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