Trust Assessment
latex-posters 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 4 findings: 0 critical, 2 high, 1 medium, and 1 low severity. Key findings include Dangerous tool allowed: Bash, Network egress to untrusted endpoints, Covert behavior / concealment directives.
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 12, 2026 (commit 458b1186). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings4
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| HIGH | Dangerous tool allowed: Bash The skill allows the 'Bash' tool without constraints. This grants arbitrary command execution. Remove unconstrained shell/exec tools from allowed-tools, or add specific command constraints. | Static | cli-tool/components/skills/scientific/latex-posters/SKILL.md:1 | |
| HIGH | Potential Command Injection via PDF Parsing Utilities in Review Script The skill declares `Bash` permission in its manifest and provides a `review_poster.sh` script. This script uses external PDF parsing utilities such as `pdfinfo`, `pdffonts`, and `pdfimages` (typically from `poppler-utils`) to analyze PDF files. While the script correctly quotes the input filename (`"$POSTER_FILE"`) to prevent direct shell injection, these underlying PDF parsing tools have historically been targets for vulnerabilities, including arbitrary code execution (RCE) when processing maliciously crafted PDF files. If an AI agent is instructed to use this skill to review an untrusted or attacker-controlled PDF, an exploit in the `poppler-utils` libraries could lead to RCE on the host system. Although the `SKILL.md` suggests reviewing 'generated' PDFs, an agent might be prompted to review any arbitrary PDF, creating a credible exploit path. 1. **Strict Input Validation**: Implement robust checks to ensure the AI agent only processes PDF files that it has generated itself or that originate from explicitly trusted sources. Avoid processing arbitrary, untrusted PDFs with this script. 2. **Sandboxing**: Execute the `review_poster.sh` script and its underlying PDF parsing utilities within a highly restricted and isolated sandbox environment (e.g., Docker container, `firejail`) to limit the potential impact of any RCE exploits. 3. **Dependency Management**: Ensure that `poppler-utils` and all other external dependencies are regularly updated to their latest versions to incorporate security patches for known vulnerabilities. 4. **Clarify Scope**: Explicitly state in the skill's documentation and internal agent instructions that the review script is intended *only* for PDFs generated by the skill itself, and not for arbitrary untrusted PDFs. | LLM | scripts/review_poster.sh:20 | |
| MEDIUM | Network egress to untrusted endpoints HTTP request to raw IP address Review all outbound network calls. Remove connections to webhook collectors, paste sites, and raw IP addresses. Legitimate API calls should use well-known service domains. | Manifest | cli-tool/components/mcps/devtools/figma-dev-mode.json:4 | |
| LOW | Covert behavior / concealment directives Multiple zero-width characters (stealth text) Remove hidden instructions, zero-width characters, and bidirectional overrides. Skill instructions should be fully visible and transparent to users. | Manifest | cli-tool/components/mcps/devtools/jfrog.json:4 |
Scan History
Embed Code
[](https://skillshield.io/report/5071ffb91cc3511c)
Powered by SkillShield