Trust Assessment
jina-reader received a trust score of 65/100, 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: 2 critical, 0 high, 0 medium, and 0 low severity. Key findings include Arbitrary command injection via header options, Arbitrary command injection via URL in 'read' mode.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 40/100, indicating areas for improvement.
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 | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command injection via header options User-controlled input for `--selector`, `--wait`, `--remove`, `--proxy`, and `--format` flags is directly interpolated into `curl` command arguments within the `HEADERS` array. An attacker can inject shell metacharacters (e.g., `";`) to execute arbitrary commands on the host system, or inject additional `curl` options (e.g., `-H "X-Malicious: injected"`) to manipulate the outgoing request. This allows for arbitrary code execution. Implement strict input validation for all user-supplied arguments (`SELECTOR`, `WAIT_SELECTOR`, `REMOVE_SELECTOR`, `PROXY_COUNTRY`, `FORMAT`) to ensure they do not contain shell metacharacters or characters that could terminate quoted strings or inject new `curl` arguments. For example, restrict values to allowed character sets or use a whitelist of acceptable values where applicable (e.g., for `FORMAT` and `PROXY_COUNTRY`). | LLM | scripts/reader.sh:50 | |
| CRITICAL | Arbitrary command injection via URL in 'read' mode The `INPUT` variable, which can contain a user-supplied URL, is directly interpolated into the `curl` command's URL argument in `read` mode. An attacker can inject shell metacharacters (e.g., `";`) into the URL to execute arbitrary commands on the host system, or inject additional `curl` options (e.g., `-X POST -d "malicious"`) to manipulate the outgoing request. This allows for arbitrary code execution. Implement strict URL validation for the `INPUT` variable in `read` mode to ensure it only contains valid URL characters and does not contain shell metacharacters or characters that could terminate quoted strings or inject `curl` options. A robust regex pattern should be used to validate the URL structure before it is used in the `curl` command. | LLM | scripts/reader.sh:72 |
Scan History
Embed Code
[](https://skillshield.io/report/8e28acfd5001043b)
Powered by SkillShield