Trust Assessment
social-media-carousel received a trust score of 10/100, placing it in the Untrusted category. This skill has significant security findings that require attention before use in production.
SkillShield's automated analysis identified 10 findings: 3 critical, 1 high, 0 medium, and 6 low severity. Key findings include Covert behavior / concealment directives, Arbitrary command execution, Remote code execution: curl/wget pipe to shell.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The LLM Behavioral Safety layer scored lowest at 53/100, 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 Findings10
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary command execution Remote code download piped to interpreter Review all shell execution calls. Ensure commands are static (not built from user input), use absolute paths, and are strictly necessary. Prefer library APIs over shell commands. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:9 | |
| CRITICAL | Remote code execution: curl/wget pipe to shell Detected a pattern that downloads and immediately executes remote code. This is a primary malware delivery vector. Never pipe curl/wget output directly to a shell interpreter. | Static | skills/okaris/social-media-carousel-2/SKILL.md:9 | |
| CRITICAL | Unsafe Remote Script Execution in Quick Start The 'Quick Start' section of the SKILL.md instructs users to install the `inference.sh` CLI by piping `curl` output directly to `sh` (`curl -fsSL https://cli.inference.sh | sh`). This method downloads and executes an unverified script from a remote server without integrity checks or version pinning. This poses a critical supply chain risk, as a compromise of `cli.inference.sh` could lead to arbitrary code execution on the user's system if they follow these instructions. Recommend a safer installation method, such as using a package manager, providing a signed installer, or requiring users to review and execute a locally downloaded script with checksum verification. | LLM | SKILL.md:8 | |
| HIGH | Unpinned External Dependencies via `npx` The skill's documentation suggests using `npx skills add inferencesh/skills@ai-image-generation` and similar commands. While `npx` is a legitimate tool, executing packages without specific version pinning (e.g., `inferencesh/skills@latest` or without a specific version) introduces a supply chain risk. If the `inferencesh/skills` package or its dependencies are compromised, a user following these instructions could execute malicious code. Always specify exact, pinned versions for `npx` packages (e.g., `inferencesh/skills@1.2.3`) to ensure reproducibility and mitigate risks from upstream package changes or compromises. | LLM | SKILL.md:191 | |
| LOW | Covert behavior / concealment directives CSS-based text hiding Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:13 | |
| LOW | Covert behavior / concealment directives CSS-based text hiding Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:54 | |
| LOW | Covert behavior / concealment directives CSS-based text hiding Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:65 | |
| LOW | Covert behavior / concealment directives CSS-based text hiding Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:74 | |
| LOW | Covert behavior / concealment directives CSS-based text hiding Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | skills/okaris/social-media-carousel-2/SKILL.md:171 | |
| LOW | Broad `Bash` Permission for `infsh` The skill declares `Bash(infsh *)` as an allowed tool. While the examples in `SKILL.md` show specific `infsh` subcommands (`app run`), the `*` wildcard allows any arguments to `infsh`. This broad permission could potentially be abused if `infsh` itself has vulnerabilities that allow arbitrary command execution through its arguments, or if user input is not properly sanitized before being passed to `infsh`. This is a potential, rather than direct, vulnerability based on the provided context. Restrict `Bash` permissions to specific `infsh` subcommands and argument patterns (e.g., `Bash(infsh app run *)`) to minimize the attack surface. Implement robust input validation and sanitization for any user-provided data passed to `infsh` commands. | LLM | Manifest:3 |
Scan History
Embed Code
[](https://skillshield.io/report/5d0bb692f326f5a5)
Powered by SkillShield