Security Audit
Sounder25/Google-Antigravity-Skills-Library:14_detect_duplicate_files
github.com/Sounder25/Google-Antigravity-Skills-LibraryTrust Assessment
Sounder25/Google-Antigravity-Skills-Library:14_detect_duplicate_files received a trust score of 78/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 3 findings: 0 critical, 1 high, 2 medium, and 0 low severity. Key findings include Potential for excessive filesystem read permissions via `--workspace-path`, Potential command injection vulnerability in `find_duplicates.ps1`, Potential for sensitive file path exfiltration via `DUPLICATE_REPORT.md`.
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 28, 2026 (commit 09376edc). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Potential for excessive filesystem read permissions via `--workspace-path` The skill's `find_duplicates.ps1` script is described as recursing through a directory specified by the `--workspace-path` parameter. If this parameter is not strictly confined to the intended workspace by the execution environment or the script itself, it could allow the skill to read arbitrary directories on the host system, leading to excessive filesystem read permissions. This broad access could expose sensitive files or configurations. Ensure the execution environment or the `find_duplicates.ps1` script strictly validates and confines the `--workspace-path` to the intended workspace or a sandboxed directory. Implement robust path sanitization and validation to prevent directory traversal attacks. | LLM | SKILL.md:25 | |
| MEDIUM | Potential command injection vulnerability in `find_duplicates.ps1` The skill relies on a `find_duplicates.ps1` script that takes string inputs like `--workspace-path` and `--exclude`. If these inputs are used to construct shell commands or file paths within the PowerShell script without proper sanitization or escaping, an attacker could inject arbitrary commands. This could lead to remote code execution or other unauthorized actions on the host system. Implement strict input validation and sanitization for all parameters passed to `find_duplicates.ps1`. When constructing commands or file paths within the script, use appropriate escaping mechanisms (e.g., `[System.IO.Path]::GetFullPath` and proper quoting) to prevent command injection. | LLM | SKILL.md:25 | |
| MEDIUM | Potential for sensitive file path exfiltration via `DUPLICATE_REPORT.md` If the `--workspace-path` parameter is exploited to scan sensitive directories (e.g., due to the excessive permissions vulnerability), the generated `DUPLICATE_REPORT.md` will contain the full paths of files found in those sensitive locations. This could lead to the exfiltration of sensitive metadata about the system's file structure, even if the file contents themselves are not exfiltrated. Mitigate the excessive permissions vulnerability by strictly confining the scan path. Additionally, consider redacting or sanitizing sensitive path components in the `DUPLICATE_REPORT.md` if the skill is intended for broader use where path information itself could be sensitive. | LLM | SKILL.md:40 |
Scan History
Embed Code
[](https://skillshield.io/report/4b86fb4077461397)
Powered by SkillShield