evolve

evolve is a command for coding agents from studioKjm/ai-harness-template. It costs 47 tokens per session (1,059 once invoked), scanned A, original, MIT.

A command that revises a system after an evaluation fails by examining the failure from several viewpoints and starting another cycle of changes.

In plain words
What is it for?
Use it after /evaluate reports a failure. It coordinates contrarian, simplifier, and researcher analyses, then repeats the process until the system converges or reaches its limit.
Why use it?
It provides a structured way to learn from failed evaluations instead of making isolated guesses about what to change.

Command

Part of the harness plugin — 44 commands, 11 agents shipped together

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 commands/studiokjm/ai-harness-template/evolve
Clone the repo
git clone --depth 1 https://github.com/studioKjm/ai-harness-template

Or install harness, the plugin that ships this one along with the rest of its 44 commands, 11 agents.

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.

agentmods badge for evolve

README.md
[![agentmods](https://agentmods.dev/badge/commands/studiokjm/ai-harness-template/evolve.svg)](https://agentmods.dev/commands/studiokjm/ai-harness-template/evolve)
Your own site
<a href="https://agentmods.dev/commands/studiokjm/ai-harness-template/evolve"><img src="https://agentmods.dev/badge/commands/studiokjm/ai-harness-template/evolve.svg" alt="Measured on agentmods" height="20"></a>
Per session 47 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,059 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 $0.00047 $0.01059
Opus 5 $0.00023 $0.00530
Sonnet 5 $0.00009 $0.00212
Haiku 4.5 $0.00005 $0.00106

Measured 5d ago against content hash dc8f74015a7d, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evolve 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 5d 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.

commands/evolve.md · 126 lines

How it starts

The opening of the file, as written. The whole thing — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/evolve — Evolution Loop

평가 결과를 반영하여 다음 세대로 진화한다

Instructions

You are the Evolver. Your job is to analyze evaluation failures and evolve the system.

Prerequisites

  • Evaluation results must exist in .harness/ouroboros/evaluations/
  • If no evaluation, prompt to run /evaluate first

Subagent Delegation (Fan-out)

진화 단계에서는 다관점 분석을 위해 subagent를 병렬로 활용합니다:

Main Agent (Evolver)
  ├─ Subagent(Contrarian)  → "이 실패가 정말 문제인가? 반대 관점은?"  (병렬)
  ├─ Subagent(Simplifier)  → "불필요한 복잡성이 원인은 아닌가?"     (병렬)
  ├─ Subagent(Researcher)  → "유사 사례/증거가 있는가?"            (병렬)
  └─ Main                  → 3개 관점 종합 → 규칙 진화 / Re-seed 결정

Claude Code에서: Agent 도구로 3개 subagent를 한 번에 병렬 spawn하여 시간 절약

Evolution Cycle

┌─────────────────────────────────────────────┐
│                                             │
│  Evaluate → Wonder → Reflect → Re-seed     │
│     ↑                            │          │
│     └────────────────────────────┘          │
│                                             │
│  Stop when: ontology similarity >= 0.95     │
│         or: max 5 generations               │
└─────────────────────────────────────────────┘

Phase 1: Wonder ("우리가 아직 모르는 것은?")

Read the latest evaluation and ask:

  1. 실패한 AC가 있나? → 왜 실패했나?
  2. 드리프트가 발생했나? → 스펙이 잘못됐나, 구현이 잘못됐나?
  3. 예상 못 한 발견이 있었나?
  4. 스펙에 빠진 것이 있었나?

Phase 2: Reflect ("피드백을 다음에 반영")

  1. Gate Evolution (Harness 연동):

    • 새로운 규칙 위반이 발견됐으면 → .harness/gates/rules/ 에 규칙 추가
    • 기존 규칙이 너무 strict했으면 → 규칙 완화
    • 새로운 패턴이 발견됐으면 → boundaries.yaml에 추가
  2. Spec Evolution:

    • 빠진 AC 추가
    • 잘못된 ontology 수정
    • constraint 조정
  3. Convention Evolution:

    • 반복된 실수 → docs/code-convention.yaml에 규칙 추가
    • 새로운 패턴 → 컨벤션으로 등록

Phase 3: Re-seed (선택적)

If the spec needs significant changes:

  • Create seed-v{N+1}.yaml with learnings incorporated
  • Reference previous version
  • Document what changed and why

Convergence Check

Compare current ontology with previous generation:

Read the full file on GitHub · 126 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. 5d ago First seen · 126 lines · 47 tokens per session scan A dc8f74015a7d

Subscribe to this mod's changes

evolve is a command published in the GitHub repository studioKjm/ai-harness-template (43 stars, last pushed 3mo ago), licensed MIT. It adds 47 tokens to every session and 1,059 once invoked, about $0.0002 per session on Opus 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 2026-08-30.