Trust Assessment
database-design received a trust score of 79/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 4 findings: 0 critical, 0 high, 3 medium, and 1 low severity. Key findings include Network egress to untrusted endpoints, Covert behavior / concealment directives, Unused and overly broad 'Write' and 'Edit' permissions declared.
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 12, 2026 (commit 458b1186). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| MEDIUM | Network egress to untrusted endpoints HTTP request to raw IP address Review all outbound network calls. Remove connections to webhook collectors, paste sites, and raw IP addresses. Legitimate API calls should use well-known service domains. | Manifest | cli-tool/components/mcps/devtools/figma-dev-mode.json:4 | |
| MEDIUM | Unused and overly broad 'Write' and 'Edit' permissions declared The skill declares 'Write' and 'Edit' permissions in its manifest, but these permissions are not utilized by the provided 'scripts/schema_validator.py' or explicitly instructed for use in 'SKILL.md'. Declaring unused broad permissions increases the attack surface if the agent were to be compromised or misused, as it grants capabilities beyond the skill's apparent needs. Remove 'Write' and 'Edit' from the 'allowed-tools' in the manifest if they are not strictly required for the skill's functionality. | LLM | SKILL.md | |
| MEDIUM | 'Glob' and 'Read' permissions can be leveraged for broad file system access via user-controlled input The skill declares 'Glob' and 'Read' permissions. The 'scripts/schema_validator.py' script, provided as a supporting file, takes a 'project_path' argument from 'sys.argv' and uses it to perform 'glob' operations ('project_path.glob(...)') and 'read_text' operations ('file_path.read_text(...)'). If the agent were to invoke this script with a 'project_path' derived from untrusted user input, it could lead to the reading and listing of arbitrary files within the agent's accessible file system. While the 'SKILL.md' does not explicitly instruct the execution of this script, the capability exists within the skill package to perform such actions. If 'project_path' is intended to be user-controlled, implement strict validation and sanitization of the path to prevent directory traversal or access to unintended locations. Consider using a more restricted file system scope if possible. If the script is only for internal use with trusted paths, ensure the agent is instructed not to pass untrusted input. | LLM | scripts/schema_validator.py:40 | |
| LOW | Covert behavior / concealment directives Multiple zero-width characters (stealth text) Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | cli-tool/components/mcps/devtools/jfrog.json:4 |
Scan History
Embed Code
[](https://skillshield.io/report/942d57717911c6aa)
Powered by SkillShield