bun-http-server
Pass
Audited by Gen Agent Trust Hub on Apr 3, 2026
Risk Level: SAFEDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [DATA_EXFILTRATION]: The implementation example for serving static files in
SKILL.mdis vulnerable to path traversal attacks. The codeconst filePath =./public${url.pathname.replace("/static", "")};allows an attacker to provide directory traversal sequences (such as..) in the URL to access files outside the intended./publicdirectory. - [PROMPT_INJECTION]: The skill documents methods for ingesting and processing untrusted data from HTTP requests, creating a vulnerability surface for indirect prompt injection.
- Ingestion points: The
req.json(),req.formData(), andreq.text()methods inSKILL.mdare shown reading external data directly into the application context. - Boundary markers: Absent; the code snippets do not demonstrate the use of delimiters or instructions for the agent to ignore embedded commands within the processed data.
- Capability inventory: The skill documentation leverages
Bun.servefor server creation andBun.filefor file system interactions. - Sanitization: The examples do not include input validation, escaping, or sanitization logic for the data received from requests.
Audit Metadata