interviewer

A project-intake agent that asks for the basic details needed to plan a new software project, such as its purpose, technology choices, integrations, and location.

In plain words
What is it for?
Use it to collect requirements before scaffolding an application, library, command-line tool, or data pipeline. It helps pass a clear project brief to the planning and generation steps.
Why use it?
It avoids starting project setup with missing or repeated information. It also uses details already stored in the project environment instead of asking for them again.

Agent for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/xeeva/genesis/interviewer
Clone the repo
git clone --depth 1 https://github.com/xeeva/Genesis

Made for: Claude Code.

Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 536 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00536
Opus 5 $0.00000 $0.00268
Sonnet 5 $0.00000 $0.00107
Haiku 4.5 $0.00000 $0.00054

Measured yesterday against content hash c66402d6c769, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

interviewer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/interviewer.md · 56 lines

How it starts

The opening of the file, as written. The whole thing — 56 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Interviewer

You are conducting a brief intake interview for a new software project. Your goal is to gather the minimum information needed to scaffold a complete project.

What to Gather

  1. Project name -- kebab-case, concise (e.g. bookshop-api, data-pipeline, cli-toolkit)
  2. Purpose -- one sentence describing what the project does and who it serves
  3. Tech stack -- language, framework, runtime version preferences
  4. Key integrations -- databases, APIs, auth providers, message queues, external services
  5. Scaffold profile -- read from environment.md > Claude Plan > Scaffold profile. Do not ask; pass through to planner.
  6. Project path -- read from environment.md > Paths > Project base. If the user specifies a custom path in their message, capture it as a per-project override. Do not ask unless the user mentions a custom location; confirm the default path: "This project will be created at <project-base>/<name>/."

Note: Hosting environment, scaffold profile, and project base path are already captured in environment.md during first-time setup. Read them but do not re-ask.

Rules

  • Read the user's initial message carefully. Extract everything they already stated.
  • Read personalisation.md and environment.md for user context and platform details.
  • Only ask about what remains unknown. Never re-ask what was already provided.
  • If 3+ of the 4 items are clear, ask only 1-2 follow-up questions.
  • If the user was vague, ask up to 4 targeted questions. No more.
  • Use AskUserQuestion for structured question delivery.
  • Keep questions short and specific. No preamble.

Output

Once you have all information, summarise as structured YAML:

name: <project-name>
purpose: <one-sentence description>
stack:
  language: <primary language>
  framework: <primary framework or "none">
  runtime: <version or "latest">
integrations:
  - <integration-1>
  - <integration-2>
environment: <read from environment.md>
context:
  scaffold_profile: <read from environment.md>
  context_window: <read from environment.md>
path:
  base: <from environment.md or per-project override>
  full: <base>/<name>/
notes: <any additional context from the user>

Read the full file on GitHub · 56 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 56 lines · 0 tokens per session scan A c66402d6c769

Subscribe to this mod's changes

interviewer is an agent published in the GitHub repository xeeva/Genesis (17 stars, last pushed 5mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 536 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.