adhoc

Execute and manage ad-hoc AI prompts in sgai workspaces via opencode commands.

2|Updated Jan 27, 2026
One-click install
npx skills add https://github.com/ucirello/sgai --skill adhoc-ucirello
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: adhoc
Source: https://github.com/ucirello/sgai/tree/main/docs/sgai-skills/adhoc
Command: npx skills add https://github.com/ucirello/sgai --skill adhoc-ucirello

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill allows you to execute single, ad-hoc AI prompts against your workspace without initiating a full, multi-agent session, streamlining quick tasks and queries.

Core Features & Use Cases

  • One-off Prompt Execution: Run any AI prompt as a standalone task.
  • Status Monitoring: Check if an ad-hoc prompt is currently running and view its output.
  • Prompt Termination: Stop a running ad-hoc prompt if needed.
  • Use Case: Quickly ask an AI to review a specific code file for potential bugs or to summarize a newly added GOAL.md file.

Quick Start

Start an ad-hoc prompt to review the authentication code and identify any security issues.

Frequently Asked Questions about adhoc

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a one-off AI prompt against my workspace without starting a full agent session?

You can run an ad-hoc AI prompt as a standalone task using non-interactive opencode commands. This executes a single query against your workspace directory, bypassing the need for a multi-agent session.

What is an ad-hoc prompt used for in an AI workspace?

An ad-hoc prompt is used for quick tasks and queries, such as asking an AI to review a specific code file for bugs or summarize a newly added GOAL.md file instantly.

Can I check the status of a running AI prompt or stop it mid-execution?

Yes, ad-hoc prompt execution supports status monitoring to check if a prompt is currently running and view its output, as well as prompt termination to stop it if needed.

Do I need a running sgai server and opencode CLI to execute ad-hoc prompts?

Yes, executing ad-hoc AI prompts requires a running sgai server and the opencode CLI to handle prompt execution and model interaction within the workspace directory.

What is the best way to summarize a GOAL.md file using AI without a multi-agent setup?

The best way is to use ad-hoc prompt execution to run a single summarization query against the file. This streamlines quick tasks without initiating a full multi-agent session.