Interviews users to scope agent ideas and generates a validated build sheet for development.
- GitHub stars
- 29,625
- Files
- 5
- Last commit
- Nov 11, 2026
- License
- MIT
- Allowed tools
- —
What does the interview skill do?
interview interviews users to scope agent ideas and generates a validated build sheet for development.
Summary generated from the skill's SKILL.md.
About the interview skill
Scoping new agent ideas through structured user interviews is the primary function of this skill. It is designed for founders or developers who need to define a single core job for an agent before starting the actual build process. By asking specific questions about triggers, inputs, actions, and success rubrics, the tool captures the user's exact requirements without inventing details.
An agent should load this skill when a user requests help scoping an agent idea. The process moves from a conversational interview to a technical plan using interview_planner.py, which maps answers to primitives and deferrals. The workflow concludes by using build_sheet_builder.py to assemble a build-sheet.json file and primitives_validator.py to ensure the plan adheres to specific limits. This approach produces a validated build sheet that serves as a blueprint for development, ensuring that version zero focuses strictly on the core task while deferring complex integrations to later phases.
What an agent can do with it
- 1Define an agent's core jobAsk the interviewer to help identify the single most important task the agent must perform end-to-end.
- 2Establish success metricsDetermine the specific rubric that will be used to grade a successful agent run.
Written from the skill's SKILL.md. Every use case is checked against a quote from it.
How to install the interview skill
Each command copies the skill folder into the client's skills directory. Restart the client or start a new session to load it.Install the interview skill in Plugin
/plugin marketplace add alirezarezvani/claude-skills /plugin install agent-launcher-skills@claude-code-skills
Install the interview skill in Claude Code
~/.claude/skills/interview/SKILL.md
$ npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview ~/.claude/skills/interviewInstall the interview skill in Claude Code (project)
.claude/skills/interview/SKILL.md
$ npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview .claude/skills/interviewInstall the interview skill in Codex
~/.agents/skills/interview/SKILL.md
$ npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview ~/.agents/skills/interviewInstall the interview skill in Gemini CLI
~/.gemini/skills/interview/SKILL.md
$ npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview ~/.gemini/skills/interviewInstall the interview skill in Claude.ai
npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview interview && zip -r interview.zip interview # claude.ai → Customize → Skills → + → Create skill → Upload a skill → interview.zip
Automated checks
Ran Nov 15, 2026Full report →Passed our automated checks. That is not a security audit: read the scripts before you let an agent run them. How checks work
- Source repository linked and reachablealirezarezvani/claude-skills/agent-launcher/skills/interview is public and reachablegithub.com/alirezarezvani/claude-skills/tree/HEAD/agent-laun ↗Pass
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit 2026-11-11github.com/alirezarezvani/claude-skills/commits ↗Pass
- No download-and-run or credential theft in scripts3 scripts read; no download-and-run, remote eval, base64 execution or credential uploadPass
- No hidden instructions in SKILL.mdNo override phrases, safeguard switches, credential requests or hidden charactersgithub.com/alirezarezvani/claude-skills/blob/HEAD/agent-laun ↗Pass
- Pre-approved tools are scopedNo pre-approved tools; the agent asks before using onegithub.com/alirezarezvani/claude-skills/blob/HEAD/agent-laun ↗Pass
- SKILL.md follows the Agent Skills formatValid name and description; says when to use it; 85 linesgithub.com/alirezarezvani/claude-skills/blob/HEAD/agent-laun ↗Pass
SKILL.md
View on GitHubPhase 1 — Interview → Plan
Open warmly with one or two examples from
../../references/examples-bank.md, then interview
the founder into a build sheet. No API key needed in this phase — the output
is a plan.
The six intake slots (ask one at a time; use AskUserQuestion for choices)
| Slot | Question | Maps to |
|---|---|---|
| Job | "What one job should this agent do end-to-end?" | agent.system + outcome description |
| Trigger | "What kicks it off — you ask it, an event, or a schedule?" | on-demand / event / cron |
| Inputs | "What does it read?" (files, repo, memory, gmail/slack/github, web) | resources / MCP servers / memory |
| Actions | "What does it do?" (draft, write, call APIs, run code) | agent toolset / custom tools / MCP |
| Done | "How would you grade a good run?" | outcome rubric (required) |
| Recurrence | "Once, on request, or on a cadence?" | single-pass / grade-loop / cron-loop |
See ../../references/interview-to-config.md
for the full mapping.
Workflow
- Interview. Walk the six slots. Capture the founder's own words — never invent specifics they didn't claim.
- Map to primitives.
MCP inputs become schema-true mock custom tools in v0 and a v1 deferral to wire the real server. Irreversible actions (send/publish) become v2 deferrals behindpython3 scripts/interview_planner.py \ --job "Triage overnight support email" --trigger schedule \ --inputs "gmail,memory" --actions "label,reply" \ --dod "one label per email, grounded reason, no invented facts" \ --recurrence daily --out ./my-agent/plan.jsonalways_ask. - Assemble the sheet.
python3 scripts/build_sheet_builder.py --plan ./my-agent/plan.json --out-dir ./my-agent - Validate limits.
FAIL blocks progress; fix and re-run. WARN is advisory (surface it).python3 scripts/primitives_validator.py --sheet ./my-agent/build-sheet.json - Record the plan in the goal.
goal_state.py set --phase stage-launch --artifact build_sheet=./my-agent/build-sheet.json, then advance.
Hard rules
- v0 is the core job only. Everything else is a versioned deferral with a reason and an exact mechanism.
- Their problem, their words.
- No key yet. The interview produces a plan; the key is a Phase-2 concern.
Forcing-question library (recommend + cite)
- "What one job — singular?" Recommend: the most-repeated task. Cite: interview-to-config.md. Two jobs → two agents.
- "Real integration or v0 mock?" Recommend: mock; wire MCP as v1. Cite: interview-to-config.md rule 1.
- "How do you grade it?" Recommend: 3–5 grounded rubric lines. Cite: cma-primitives.md (rubric required).
- "Smarter over time?" Recommend: attach memory only if yes. Cite: cma-primitives.md (memory limits + injection).
- "Once, or on a cadence?" Recommend: on-demand v0, schedule as Phase-4. Cite: loops-and-workflows.md.
Tools
scripts/interview_planner.py— answers → primitives skeleton + deferrals.scripts/build_sheet_builder.py— assemble/normalize build-sheet.json.scripts/primitives_validator.py— validate vs CMA limits (PASS/WARN/FAIL).
Files in this skill
5interview skill questions
How do I install the interview skill in Claude Code?
Run npx degit alirezarezvani/claude-skills/agent-launcher/skills/interview ~/.claude/skills/interview. Claude Code loads skills from ~/.claude/skills (all projects) or .claude/skills (one project) and uses a skill when its description matches the task.
Which clients can use the interview skill?
Any client that reads the Agent Skills format: Claude Code, Codex, Gemini CLI and claude.ai (upload as a ZIP). Scripts in the skill run only where the client allows code execution.
Do I need an API key to use this skill for scoping?
No, an API key is not required during the interview phase as the output is a plan rather than a live agent.
How is the generated build sheet validated?
The build sheet is checked using primitives_validator.py to ensure it adheres to CMA limits, returning PASS, WARN, or FAIL statuses.