bun-deploy
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process local project files (e.g., TypeScript source code, package.json, bun.lockb) to perform build and deployment operations. This creates a surface for indirect prompt injection if the project being processed is maliciously crafted. \n
- Ingestion points: Processes files in the current working directory, including ./src/cli.ts, ./src/index.ts, package.json, and bun.lockb as seen in SKILL.md and references/docker.md. \n
- Boundary markers: There are no explicit instructions to the agent to treat these files as untrusted or to ignore embedded instructions within comments or metadata. \n
- Capability inventory: The skill utilizes commands such as bun install, bun build, and docker build, which may execute arbitrary code or scripts defined in configuration files (e.g., lifecycle scripts in package.json or RUN instructions in a Dockerfile). \n
- Sanitization: No sanitization or validation of input file content is performed before processing.
Audit Metadata