convert-farcaster-miniapp-to-app
Pass
Audited by Gen Agent Trust Hub on Mar 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands including
npm install,npx tsc --noEmit, andnpm run buildwithin the user's project directory. This represents a risk if the project being converted contains malicious scripts inpackage.jsonor complex build configurations. - [EXTERNAL_DOWNLOADS]: The skill triggers external downloads through
npm installand provides code snippets designed to fetch data from Farcaster hubs (hub.farcaster.xyz). - [PROMPT_INJECTION]: There is a surface for indirect prompt injection because the skill reads and processes untrusted project files (source code and configuration) to determine its migration logic. Maliciously crafted comments or code patterns in the analyzed project could theoretically influence the agent's behavior during the conversion process.
- Ingestion points: Reads
package.json, source files (.ts,.tsx,.js,.jsx), and.envfiles from the local project. - Boundary markers: None explicitly defined for the data processed during analysis or conversion phases.
- Capability inventory: File read/write, shell command execution (
npm,npx), and network requests (via generated code). - Sanitization: No specific sanitization or filtering of input code is mentioned before the agent performs transformations.
Audit Metadata