Security Audit
seo-fundamentals
github.com/sickn33/antigravity-awesome-skillsTrust Assessment
seo-fundamentals received a trust score of 85/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 1 finding: 0 critical, 0 high, 1 medium, and 0 low severity. Key findings include Potential Data Exposure via Local File Content Reading.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 20, 2026 (commit e36d6fd3). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Potential Data Exposure via Local File Content Reading The `seo_checker.py` script is designed to read content from files within a user-specified `project_path`. While it includes safeguards like skipping common sensitive directories (`SKIP_DIRS`) and filtering files based on `SKIP_PATTERNS`, its `is_page_file` logic is broad. It identifies files as 'pages' based on extensions (`.html`, `.jsx`, `.tsx`), directory names (`pages`, `app`, `routes`), or common page filenames (`index`, `home`). This broad criteria could lead to the script reading internal or sensitive files (e.g., `src/components/admin/dashboard/index.jsx` or `internal/docs/index.html`) if they match the 'page file' heuristics and are not located in a skipped directory. The content of these files is then processed by the script and its analysis (which implicitly relies on the content) is printed to standard output, making it accessible to the host LLM. This poses a risk of unintended data exposure or exfiltration if the LLM is subsequently prompted to process or transmit this data. 1. **Refine `is_page_file` logic:** Implement more stringent criteria for identifying public-facing pages, potentially requiring explicit configuration or stricter path validation. 2. **Strict sandboxing:** Ensure the agent's execution environment strictly limits filesystem access to only explicitly allowed, non-sensitive directories. 3. **User confirmation:** For skills that read local files, consider requiring explicit user confirmation for each file or directory access, especially for non-public paths. | LLM | scripts/seo_checker.py:90 |
Scan History
Embed Code
[](https://skillshield.io/report/ee9268513417395f)
Powered by SkillShield