What is evolve-interview?

A guided interview that turns a vague evolutionary coding task into precise files and settings. Evolutionary code discovery repeatedly changes code and measures whether each version improves.

Use it to define the goal, editable code, fitness measurement, and constraints, producing initial.py, evaluate.py, and config.json.

It prevents expensive runs from optimizing the wrong code or using an unclear or unreliable success test. The interview continues until the task is mathematically specific enough to evaluate.

evolve-interview — Skill for Claude Code by samuelzxu/claude-evolve

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.

Any agent

npx skills add samuelzxu/claude-evolve --skill evolve-interview

Clone the repo

git clone --depth 1 https://github.com/samuelzxu/claude-evolve

Made for: Claude Code.

Or install claude-evolve, the plugin that ships this one along with the rest of its 4 skills, 1 MCP server.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/samuelzxu/claude-evolve/evolve-interview/github.svg)](https://agentmods.dev/skills/samuelzxu/claude-evolve/evolve-interview)

Your own site

<a href="https://agentmods.dev/skills/samuelzxu/claude-evolve/evolve-interview"><img src="https://agentmods.dev/badge/skills/samuelzxu/claude-evolve/evolve-interview/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/samuelzxu/claude-evolve/evolve-interview"><img src="https://agentmods.dev/badge/skills/samuelzxu/claude-evolve/evolve-interview.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 31 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 4,487 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.00031	$0.04487
Opus 5.5	$0.00012	$0.01795
Sonnet 5.5	$0.00006	$0.00897
Haiku 4.5	$0.00003	$0.00449

Measured 2mo ago against content hash 3e0454d9d45e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

evolve-interview 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 2mo ago.

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.

skills/evolve-interview/SKILL.md · 432 lines

How it starts

The opening of the file, as written. The whole thing — 432 lines — stays where its author put it; the contents beside it link to each section on GitHub.

claude-evolve Interview

Ouroboros-inspired Socratic interview that refuses to start an expensive evolution run until the task specification is mathematically clear. Asks targeted questions across four dimensions, scores ambiguity after every answer, and only crystallizes into concrete artifacts (initial.py, evaluate.py, config.json) once ambiguity drops below 20%.

When this skill is invoked, immediately execute the workflow below. Do not only restate or summarize these instructions back to the user.

Purpose

Evolutionary code discovery is expensive. Each generation burns real LLM calls and wall-clock time. If the fitness function is ambiguous, the wrong code region is marked mutable, or the evaluator is unreliable, the entire run produces garbage. This skill prevents that by forcing specification clarity before execution.

Analogous to oh-my-claudecode's deep-interview, but specialized for the four dimensions that actually matter for evolution:

Goal Clarity — What are we optimizing?
Program Clarity — Which code is mutable?
Evaluation Clarity — How is fitness measured (without LLM-as-judge)?
Constraint Clarity — What must be preserved? What's the budget?

Use When

User wants to evolve code but hasn't defined initial.py / evaluate.py yet
User has a vague optimization goal ("make this faster", "improve accuracy")
User has code but isn't sure what to mark as mutable with EVOLVE-BLOCK markers
User says "help me set up an evolution run", "interview me", "I want to evolve X"

Do Not Use When

User already has initial.py with EVOLVE-BLOCK markers AND a working evaluate.py → run /evolve directly
User just wants to check status of a running evolution → run /evolve-status
User wants to install the plugin → run /evolve-install

Execution Policy

Ask ONE question at a time — never batch
Target the weakest clarity dimension with each question
State, in one sentence before the question, why that dimension is the bottleneck
Gather codebase facts via the Explore subagent BEFORE asking the user about them
For brownfield tasks, cite repo evidence (file path, function, line) in questions
Score ambiguity after every answer and display it transparently
Do NOT proceed to crystallization until ambiguity ≤ 0.20 (or the user explicitly exits early with a warning)
Persist interview state for resume across interruptions
Challenge agent modes activate at specific round thresholds to break fixation

Read the full file on GitHub · 432 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.

2mo ago First seen · 432 lines · 31 tokens per session scan A 3e0454d9d45e

Subscribe to this mod's changes

evolve-interview is a skill published in the GitHub repository samuelzxu/claude-evolve (17 stars, last pushed 4mo ago), licensed MIT. It adds 31 tokens to every session and 4,487 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-09-01.

Building agents Planning

evolve-interview next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
evolve-interviewsamuelzxu/claude-evolve	Skill	17	4mo ago	MIT	A	31	Claude Code
imagegenopenai/codex	Skill	133k	yesterday	Apache-2.0	A	113	Codex
playwright-climicrosoft/playwright	Skill	99k	yesterday	Apache-2.0	A	19	Claude Code
local-ai-agentsmicrosoft/ai-agents-for-beginners	Skill	79k	4d ago	MIT	A	200	Claude Code, Codex
systematic-debuggingobra/superpowers	Skill	313k	3d ago	MIT	A	21	Claude Code
next-cache-components-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	95	Claude Code
next-partial-prefetching-adoptionvercel/next.js	Skill	143k	yesterday	MIT	A	103	Claude Code