First Ask
Clarify scope, deliverables, and constraints with structured human input before your agent starts implementation work.
Overview
First Ask is a journey-wide agent skill that refines task scope, deliverables, and constraints through Joyride-powered human input—usable whenever a solo builder needs to align with the developer before committing to exe
Install
npx skills add https://github.com/github/awesome-copilot --skill first-askWhat is this skill?
- Iterative refinement loop: scope, deliverables, constraints, and success criteria before action
- Human-in-the-loop via `joyride_request_human_input` for every clarification and final sign-off
- Optional repo exploration and web research to ground the plan in real project context
- Organizes work into clear sections or steps and surfaces a plan before execution
- Requires the Joyride extension as the input channel for the workflow
Adoption & trust: 8.5k installs on skills.sh; 34.6k GitHub stars; 2/3 security scanners passed (skills.sh audits).
What problem does it solve?
You handed the agent a vague request and risk wasted work because nobody nailed scope, success criteria, or constraints first.
Who is it for?
Ambiguous tasks, cross-cutting changes, or any job where one wrong assumption would mean rework.
Skip if: Fully approved specs or trivial one-liner fixes where scope is already documented and no human gate is needed.
When should I use this skill?
Before carrying out a non-trivial task when scope, deliverables, or constraints are unclear; requires Joyride extension and `joyride_request_human_input`.
What do I get? / Deliverables
You get an explicit plan the human approved via interactive prompts, then the agent carries out the task against that shared understanding.
- Refined task understanding
- Step-by-step plan shown to the human
- Executed work only after human confirms no further input
Recommended Skills
Journey fit
Useful at every journey phase - explore requirements and options before committing to a direction.
Where it fits
Interrogate what ‘research competitors’ should deliver (report format, depth, sources) before the agent crawls the web.
Lock MVP boundaries and success metrics for a landing-page experiment before any code is written.
Clarify API contracts, error handling, and migration constraints before implementing an endpoint.
Confirm rollout steps, feature flags, and rollback criteria before executing a release checklist.
Define what ‘fix production slowness’ means (SLO, metrics, blast radius) before touching infra.
How it compares
Use instead of jumping straight from chat prompt to code when alignment matters more than speed.
Common Questions / FAQ
Who is first-ask for?
Solo and indie builders using Copilot-style agents with the Joyride extension who want structured Q&A before implementation.
When should I use first-ask?
At validate when scoping a feature, at build before a large refactor, at ship when launch steps are unclear, or at operate when an incident task needs confirmed constraints—anytime the task is not fully specified.
Is first-ask safe to install?
Review the Security Audits panel on this Prism page and treat Joyride human-input and any web research the skill triggers as permissions you control in your environment.
SKILL.md
READMESKILL.md - First Ask
# Act Informed: First understand together with the human, then do You are a curious and thorough AI assistant designed to help carry out tasks with high-quality, by being properly informed. You are powered by the `joyride_request_human_input` tool and you use it as a key part of your process in gathering information about the task. <refining> Your goal is to iteratively refine your understanding of the task by: - Understanding the task scope and objectives - At all times when you need clarification on details, ask specific questions to the user using the `joyride_request_human_input` tool. - Defining expected deliverables and success criteria - Perform project explorations, using available tools, to further your understanding of the task - If something needs web research, do that - Clarifying technical and procedural requirements - Organizing the task into clear sections or steps - Ensuring your understanding of the task is as simple as it can be </refining> After refining and before carrying out the task: - Use the `joyride_request_human_input` tool to ask if the human developer has any further input. - Keep refining until the human has no further input. After gathering sufficient information, and having a clear understanding of the task: 1. Show your plan to the user with redundancy kept to a minimum 2. Create a todo list 3. Get to work!