Security Audit
sundial-org/awesome-openclaw-skills:skills/assemblyai-transcribe
github.com/sundial-org/awesome-openclaw-skillsTrust Assessment
sundial-org/awesome-openclaw-skills:skills/assemblyai-transcribe 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 3 findings: 1 critical, 2 high, 0 medium, and 0 low severity. Key findings include Sensitive path access: AI agent config, Arbitrary File Read and Exfiltration, Arbitrary File Write via --out Flag.
The analysis covered 4 layers: Manifest Analysis, Static Code Analysis, Dependency Graph, LLM Behavioral Safety. The Static Code Analysis layer scored lowest at 40/100, indicating areas for improvement.
Last analyzed on March 3, 2026 (commit 6d998e00). SkillShield performs automated 4-layer security analysis on AI skills and MCP servers.
Layer Breakdown
Behavioral Risk Signals
Security Findings3
| Severity | Finding | Layer | Location | |
|---|---|---|---|---|
| CRITICAL | Arbitrary File Read and Exfiltration The skill's `transcribe` command allows users to specify an arbitrary `filePath` for audio/video input. The `uploadFile` function resolves this path using `path.resolve(expandHome(filePath))` and then reads the file content using `fs.createReadStream(abs)`. This content is then uploaded to AssemblyAI's servers. This means a malicious user could provide a path to any file readable by the Node.js process (e.g., `/etc/passwd`, `~/.ssh/id_rsa`, `~/.aws/credentials`) and cause its contents to be exfiltrated to a third-party service (AssemblyAI). This is a severe data exfiltration risk. Implement strict input validation and sandboxing for file paths. The `filePath` argument should be restricted to a designated, temporary upload directory or require explicit user confirmation for access to specific files. The execution environment for the skill should enforce least privilege, limiting filesystem access to only necessary directories. Avoid allowing arbitrary file paths from untrusted input. | Static | assemblyai.mjs:169 | |
| HIGH | Sensitive path access: AI agent config Access to AI agent config path detected: '~/.clawdbot/'. This may indicate credential theft. Verify that access to this sensitive path is justified and declared. | Static | skills/assemblyai-transcribe/SKILL.md:19 | |
| HIGH | Arbitrary File Write via --out Flag The skill's commands (e.g., `transcribe`, `subtitles`, `paragraphs`, `sentences`, `get`) support an `--out <path>` flag, which allows writing the output to an arbitrary file path. The `writeOutput` function resolves this path using `path.resolve(expandHome(outPath))` and then writes content using `fsp.writeFile(absOutPath, output)`. This could allow a malicious user to overwrite existing files, write to sensitive system locations, or fill up disk space, leading to data corruption, denial of service, or other system integrity issues. Restrict output file paths to a designated, temporary output directory. Do not allow arbitrary file paths from untrusted input for writing. If writing to user-specified locations is necessary, implement robust checks to prevent overwriting sensitive files or writing outside of allowed directories. The execution environment should enforce least privilege for file write operations. | Static | assemblyai.mjs:310 |
Scan History
Embed Code
[](https://skillshield.io/report/e5961cd181d81e15)
Powered by SkillShield