evolve-spec

A command for reviewing the results of an automated development cycle and updating that cycle’s written capability plan.

In plain words
What is it for?
Use it to read recent results, suggest feasible additions, get approval, update the matching specification, and check that the revised plan remains consistent.
Why use it?
It helps keep the plan aligned with what was actually built and with useful follow-up ideas discovered during implementation.

Command

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/aminry/auto-claude-plugins/evolve-spec
Clone the repo
git clone --depth 1 https://github.com/aminry/auto-claude-plugins
Per session 15 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,003 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 $0.00015 $0.02003
Opus 5 $0.00008 $0.01001
Sonnet 5 $0.00003 $0.00401
Haiku 4.5 $0.00002 $0.00200

Measured 2d ago against content hash f8c4529249ba, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

evolve-spec 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 2d 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.

plugins/auto-develop/commands/evolve-spec.md · 227 lines

How it starts

The opening of the file, as written. The whole thing — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Analyze the latest auto-develop results to discover new capabilities that are now possible or valuable, then update the capability specification accordingly.

The auto-develop skill to analyze: $ARGUMENTS


Overview

After running an auto-develop cycle, the capability specification may have room to grow — new capabilities that are now possible because of what was built, natural follow-ons, or capabilities discovered during implementation. This skill:

  1. Reads the latest auto-develop results and the current spec
  2. Analyzes what was built and what new possibilities emerged
  3. Identifies new capabilities that are now valuable or feasible
  4. Proposes additions and gets user approval
  5. Updates the capability spec
  6. Validates the updated spec consistency

Phase 1: Load Context

  1. Identify the feature: Parse $ARGUMENTS to determine the auto-develop skill name. Expected format: a feature name like search, auth, export, etc.

  2. Load the current spec: Read .claude/commands/spec-{feature}.md (or the appropriate spec file — check .claude/commands/ for the matching spec).

  3. Load the results log: Read .debug/auto-develop-{feature}/results.tsv to understand the implementation trajectory — what was attempted, accepted, rejected, and in what order.

  4. Load the regression suite: Read .debug/auto-develop-{feature}/regression.json to see all currently-tracked capabilities and baseline checks.

  5. Load implementation logs: Read the latest iteration's plan and verification files from .debug/auto-develop-{feature}/logs/ and .debug/auto-develop-{feature}/snapshots/ to understand what was actually built and any issues encountered.

  6. Load the current code: Read the files that were modified during auto-develop (from git log or the implementation plans) to see the current state of the feature.

If any of these files don't exist, report what's missing and stop.


Phase 2: Analyze What Was Built

For each implemented capability ([x] status), examine:

Read the full file on GitHub · 227 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. 2d ago First seen · 227 lines · 15 tokens per session scan A f8c4529249ba

Subscribe to this mod's changes

evolve-spec is a command published in the GitHub repository aminry/auto-claude-plugins (1 stars, last pushed 3mo ago), licensed MIT. It adds 15 tokens to every session and 2,003 once invoked, about $0.0001 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 2026-08-31.