📦

Audit-Verlauf

backlink-analyzer - 2 Audits

Audit-Version 2

Neueste Niedriges Risiko

Jun 27, 2026, 06:48 PM

Static analysis reported many high-risk patterns, but review found them in Markdown examples, SEO terminology, example URLs, and relative documentation links. No executable scripts, hidden network requests, credential access, data exfiltration, or prompt injection attempts were found in the reviewed files.

Gescannte Dateien

1,468

Analysierte Zeilen

Review items

False positives ignored

Capability review items (4)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Niedrig

SKILL.md:35-55 SKILL.md:80-455 references/link-quality-rubric.md:217-230

Markdown Examples Misclassified as Shell Execution

The external command alerts are Markdown code fences and output templates. They show user prompts and report formats, not executable shell commands or Ruby backticks.

The flagged ranges are fenced Markdown examples and plain text templates. No script file, command runner, or user-input command construction is present.

Niedrig

references/link-quality-rubric.md:82 references/link-quality-rubric.md:223 references/link-quality-rubric.md:224

Example URLs Misclassified as Network Activity

The hardcoded URLs are examples in backlink and disavow documentation. They are not used by code to fetch, post, or transmit data.

The URLs appear inside explanatory Markdown and a sample disavow file. No network API, connector call, or automatic request logic surrounds them.

Niedrig

SKILL.md:59 SKILL.md:589-594

Relative Reference Links Misclassified as Path Traversal

The path traversal alerts are relative Markdown links to connector documentation and related skills. They do not read files or access arbitrary paths.

The affected lines are documentation hyperlinks. There is no file-system API, path concatenation, or user-controlled file access.

Niedrig

SEO Terms Misclassified as Blocker Patterns

The weak cryptography and reconnaissance alerts are false positives from SEO language, outreach text, and scoring tables. No cryptographic function or system probing is present.

The flagged lines are natural-language SEO guidance, email templates, or scoring rubric entries. Reviewed context showed no cryptographic code, host discovery, or reconnaissance workflow.

Risikofaktoren

⚙️ Externe Befehle (3)

references/link-quality-rubric.md:217-230 SKILL.md:35-55 SKILL.md:80-455

🌐 Netzwerkzugriff (3)

references/link-quality-rubric.md:82 references/link-quality-rubric.md:223 references/link-quality-rubric.md:224

📁 Dateisystemzugriff (2)

SKILL.md:59 SKILL.md:589-594

Geprüft von: codex

Audit-Version 1

Sicher

Feb 12, 2026, 08:56 AM

Static analysis detected 78 potential security issues across 3 files (1468 lines). After comprehensive evaluation, all findings are confirmed false positives. The detected patterns are markdown code blocks showing example commands, documentation URLs, SEO metric abbreviations (DR/DA), and relative file path references in documentation. This is a legitimate SEO analysis skill with no executable code, making it safe for marketplace publication.

Gescannte Dateien

1,468

Analysierte Zeilen

Review items

False positives ignored

Capability review items (3)

These are real local capabilities that may be expected for this skill, so they require review but are not counted as confirmed malicious behavior.

Niedrig

SKILL.md:35-55 references/link-quality-rubric.md:217-230

Markdown Code Blocks Flagged as External Commands

Static scanner flagged 32 instances of Ruby/shell backtick execution patterns. These are markdown code block delimiters (```) used for formatting documentation examples, not actual code execution. Found in SKILL.md and reference documentation showing bash command examples and email templates.

Clear evidence that backticks are markdown code block delimiters surrounding documentation examples, not executable code

Niedrig

references/link-quality-rubric.md:82 references/link-quality-rubric.md:223-224

Documentation URLs Flagged as Network Calls

Static scanner flagged 3 instances of hardcoded URLs as network security risks. These are example URLs in markdown documentation (e.g., spam-site.com in disavow file examples) and relative file path references, not actual network requests.

URLs appear in markdown documentation as examples within code blocks, not as executable network calls

Niedrig

SKILL.md:59 SKILL.md:589-594

Relative File Paths Flagged as Path Traversal

Static scanner flagged 12 instances of path traversal sequences (../). These are relative file path references in markdown documentation linking to other skill files (e.g., ../../CONNECTORS.md), not malicious path traversal attacks.

Parent directory references are standard markdown link syntax for documentation navigation, not exploit attempts

Risikofaktoren

⚙️ Externe Befehle (3)

references/link-quality-rubric.md:217-230 SKILL.md:35-55 SKILL.md:80-120

🌐 Netzwerkzugriff (2)

references/link-quality-rubric.md:82 references/link-quality-rubric.md:223-224

📁 Dateisystemzugriff (2)

SKILL.md:59 SKILL.md:589-594

Geprüft von: claude

← Zurück zum Skill