Security Audit
existential-birds/beagle:plugins/beagle-core/skills/fix-llm-artifacts
github.com/existential-birds/beagleTrust Assessment
existential-birds/beagle:plugins/beagle-core/skills/fix-llm-artifacts received a trust score of 85/100, 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 Command Injection via Untrusted JSON Fields.
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 July 20, 2026 (commit aed6fce7). 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 | Command Injection via Untrusted JSON Fields The skill reads the `target` and `scope` fields from `.beagle/llm-artifacts-review.json` (which can be modified by untrusted actors in a repository) and instructs the LLM to invoke shell commands using these variables (e.g., `review-llm-artifacts "$stored_target"`). If an attacker crafts a malicious review JSON file containing shell metacharacters in the `target` field, it can lead to arbitrary command execution when the LLM executes the command. Sanitize the `stored_target` and `stored_scope` values to ensure they only contain valid, safe path characters (e.g., alphanumeric, slashes, periods, dashes) and do not contain shell metacharacters before passing them to any shell command execution. Alternatively, avoid executing shell commands directly with untrusted inputs from the workspace. | LLM | SKILL.md:121 |
Scan History
Embed Code
[](https://skillshield.io/report/dc863e133c5b0521)
Powered by SkillShield