Trust Assessment
enhance-prompt received a trust score of 86/100, 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, 1 high, 0 medium, and 0 low severity. Key findings include User-controlled arbitrary file write via 'Custom filename' option.
The analysis covered 4 layers: dependency_graph, static_code_analysis, manifest_analysis, llm_behavioral_safety. All layers scored 70 or above, reflecting consistent security practices.
Last analyzed on February 18, 2026 (commit 51d6d5ab). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings1
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | User-controlled arbitrary file write via 'Custom filename' option The skill's 'Output Options' section explicitly states that if the user requests, the enhanced prompt can be written to a 'Custom filename specified by user'. This allows an untrusted user to provide an arbitrary file path, which could lead to path traversal vulnerabilities (e.g., `../../../../etc/passwd`). This could result in overwriting critical system files, creating new files in sensitive locations, or potentially aiding in data exfiltration if sensitive output is written to an accessible location. The 'Write' permission declared in the manifest enables this risk. 1. Restrict file paths: If custom filenames are necessary, implement strict validation to ensure paths are relative to the project root and do not contain path traversal sequences (`../`). 2. Whitelist directories: Only allow writing to a predefined, safe subdirectory (e.g., a `generated_prompts` folder). 3. Remove arbitrary file write: If not strictly necessary, remove the option for users to specify a custom filename, and only allow writing to hardcoded, safe filenames (like `next-prompt.md`). 4. Confirm intent: Ensure the LLM is instructed to *only* write the *generated prompt content* and not to execute any commands embedded in the filename. | Unknown | SKILL.md:154 |
Scan History
Embed Code
[](https://skillshield.io/report/39c0736659999604)
Powered by SkillShield