Trust Assessment
predictme received a trust score of 82/100, placing it in the Mostly Trusted category. This skill has passed most security checks with only minor considerations noted.
SkillShield's automated analysis identified 2 findings: 0 critical, 1 high, 1 medium, and 0 low severity. Key findings include API Key Stored in Local File, Filesystem Access for State Management.
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 13, 2026 (commit 13146e6a). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings2
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | API Key Stored in Local File The skill explicitly instructs the agent to save the obtained `api_key` and `agentId` to a local file, `~/.predictme/credentials.json`. Storing sensitive API keys directly in local files increases the risk of credential harvesting if the agent's host environment is compromised or if the file is not adequately protected. This practice makes the API key persistently available on the system. Store API keys in a secure secrets management system (e.g., environment variables, a dedicated secrets store, or a platform-provided secure storage mechanism) rather than directly in local files. Ensure the agent's runtime environment restricts filesystem access to only necessary paths. | LLM | skill.md:48 | |
| MEDIUM | Filesystem Access for State Management The skill requires the agent to read and write files on the local filesystem (e.g., `~/.predictme/credentials.json`, `preferences.json`, `nonce.json`) for credential storage, user preferences, and state management. This necessitates filesystem access. If the agent is granted overly broad permissions (e.g., write access to arbitrary directories), it could lead to unauthorized file manipulation or data exfiltration beyond the scope of the skill's intended operation. Implement a principle of least privilege for the agent's filesystem access. Restrict read/write operations to only the specific directories and files absolutely necessary for the skill's operation (e.g., a dedicated skill data directory), and avoid writing to user home directories or system-wide locations. For non-sensitive state like `nonce`, consider using a dedicated, sandboxed storage mechanism if available. | LLM | skill.md:59 |
Scan History
Embed Code
[](https://skillshield.io/report/7e7f0c4da36bea81)
Powered by SkillShield