Trust Assessment
ezbookkeeping received a trust score of 72/100, placing it in the Caution category. This skill has some security considerations that users should review before deployment.
SkillShield's automated analysis identified 2 findings: 0 critical, 2 high, 0 medium, and 0 low severity. Key findings include Skill enables arbitrary shell command execution with potential for injection, Unpinned script downloads from remote repository.
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 14, 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 | Skill enables arbitrary shell command execution with potential for injection The skill instructs the AI agent to download and execute external shell scripts (`ebktools.sh` and `ebktools.ps1`) using `curl` and `Invoke-WebRequest`, followed by `chmod +x` and direct execution. This grants the AI agent the ability to run arbitrary commands on the host system with the permissions of the user running the agent. The usage examples (`./ebktools.sh <command> [command-options]` and `.\ebktools.ps1 <command> [command-options]`) indicate that user-controlled input (`<command>`, `[command-options]`) will be passed directly to a shell script, creating a high-risk vector for command injection if these inputs are not rigorously sanitized by the `ebktools` scripts. This constitutes excessive permissions and a direct command injection vulnerability. Avoid direct execution of downloaded scripts. If shell execution is absolutely necessary, ensure all inputs are rigorously sanitized and consider using a sandboxed environment or a more controlled execution mechanism. Pin script versions to specific commit hashes to mitigate supply chain risks. | LLM | SKILL.md:14 | |
| HIGH | Unpinned script downloads from remote repository The skill downloads `ebktools.sh` and `ebktools.ps1` scripts from `raw.githubusercontent.com` using `refs/heads/main`. This means the scripts are not pinned to a specific version or commit hash. If the `main` branch of the `mayswind/ezbookkeeping` repository is compromised, a malicious actor could alter the scripts, leading to a supply chain attack where the AI agent downloads and executes compromised code without detection. Pin the downloaded scripts to a specific commit hash or version tag (e.g., `refs/tags/v1.0.0` or `refs/heads/<commit_hash>`) to ensure deterministic and verifiable script content. Alternatively, bundle the scripts directly with the skill package if licensing permits. | LLM | SKILL.md:14 |
Scan History
Embed Code
[](https://skillshield.io/report/e026919db2354d4c)
Powered by SkillShield