Trust Assessment
ffmpeg-master received a trust score of 86/100, 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 `ffmpeg -safe 0` usage allows arbitrary file access.
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 13, 2026 (commit 13146e6a). 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 | `ffmpeg -safe 0` usage allows arbitrary file access The skill documentation includes an example `ffmpeg` command using `-f concat -safe 0 -i filelist.txt`. The `-safe 0` option explicitly disables security checks for file paths specified within `filelist.txt`. If an LLM generates `ffmpeg` commands based on this example and allows `filelist.txt` (or its contents) to be influenced by untrusted user input, an attacker could craft `filelist.txt` to include arbitrary file paths (e.g., `/etc/passwd`, `/root/.ssh/id_rsa`). This could lead to:
1. **Data Exfiltration**: Sensitive files could be read and their contents concatenated into the output file.
2. **Command Injection (indirect)**: While not direct shell injection, it allows for reading/writing arbitrary files, which is a powerful primitive for further attacks or system compromise.
3. **Excessive Permissions**: The `-safe 0` flag grants `ffmpeg` broader file system access than typically desired for a skill processing untrusted input. Avoid using `-safe 0` with `ffmpeg -f concat` when processing untrusted input. If concatenation of multiple files is required, ensure that the `filelist.txt` content is strictly controlled and sanitized, or use a safer method for concatenation that does not involve disabling security checks. If the skill is intended to be used by an LLM, the LLM should be instructed to never generate commands with `-safe 0` or to rigorously sanitize any user-provided file paths before including them in `filelist.txt`. | LLM | SKILL.md:31 |
Scan History
Embed Code
[](https://skillshield.io/report/988659a8c7d2b448)
Powered by SkillShield