electrobun-kitchen-sink
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions for executing various shell commands such as
bun install,electrobun dev, andnpx tsx scripts/generate-manifest.ts. These commands are standard for project setup, development, and manifest generation within the described Electrobun environment. - [INDIRECT_PROMPT_INJECTION]: The skill describes workflows for reading and processing a JSON manifest file (
kitchen/src/generated/feature-manifest.json), which represents an ingestion surface for potentially untrusted data if the file is controlled by an external actor. - Ingestion points: The manifest file is read using
catand processed withjqin the 'Consuming the Manifest' section. - Boundary markers: No delimiters or instructions to ignore embedded content are specified when reading the manifest.
- Capability inventory: The skill documents the use of shell execution tools (
bun,electrobun,npx) and data processing utilities (jq). - Sanitization: No validation or sanitization of the manifest's JSON fields (like
titleordescription) is described before the agent interacts with them.
Audit Metadata