skills/tychohq/agent-skills/dev-serve/Gen Agent Trust Hub

dev-serve

Pass

Audited by Gen Agent Trust Hub on Sep 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The dev-serve.sh script extracts commands from a project's package.json file (specifically the scripts.dev field) and executes them within a tmux session. While this is the intended purpose of the tool, it involves running shell commands derived from local file content.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect injection as it ingests untrusted data from a project repository to influence its behavior.
  • Ingestion points: The tool reads the package.json and various vite.config.* files from the directory provided by the user.
  • Boundary markers: No explicit boundary markers or warnings are used when processing the content of these files.
  • Capability inventory: The skill has the capability to execute shell commands via tmux send-keys, modify local files using sed (including the Caddyfile and Vite configuration files), and perform local network requests to the Caddy admin API.
  • Sanitization: The script performs minimal sanitization on the commands extracted from package.json before execution.
  • [DYNAMIC_EXECUTION]: The tool performs dynamic modification of executable or configuration files at runtime. It uses sed to inject host entries into Vite configuration files (vite.config.ts, etc.) and appends reverse proxy blocks to the system's Caddyfile based on the folder name and detected ports.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 23, 2026, 06:38 AM