simplicio-autoresearch

simplicio-autoresearch is a skill for Claude Code, Codex from wesleysimplicio/simplicio-loop. It costs 194 tokens per session (2,473 once invoked), scanned A, original, MIT.

A controlled loop for improving a target by proposing changes, measuring them against fixed criteria, and keeping only improvements. It is based on repeated experiments with limits on iterations, time, and resources.

In plain words
What is it for?
Use it for metric-driven code or configuration experiments that can be evaluated consistently. The input says the assistant proposes mutations while a worker enforces the limits and bookkeeping.
Why use it?
It prevents open-ended optimization and records whether each change improved the measured result or should be reverted.

Skill for Claude CodeCodex

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 skills/wesleysimplicio/simplicio-loop/simplicio-autoresearch
Any agent
npx skills add wesleysimplicio/simplicio-loop --skill simplicio-autoresearch
Clone the repo
git clone --depth 1 https://github.com/wesleysimplicio/simplicio-loop

Made for: Claude Code, Codex.

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 simplicio-autoresearch

README.md
[![agentmods](https://agentmods.dev/badge/skills/wesleysimplicio/simplicio-loop/simplicio-autoresearch.svg)](https://agentmods.dev/skills/wesleysimplicio/simplicio-loop/simplicio-autoresearch)
Your own site
<a href="https://agentmods.dev/skills/wesleysimplicio/simplicio-loop/simplicio-autoresearch"><img src="https://agentmods.dev/badge/skills/wesleysimplicio/simplicio-loop/simplicio-autoresearch.svg" alt="Measured on agentmods" height="20"></a>
Per session 194 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,473 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 $0.00194 $0.02473
Opus 5 $0.00097 $0.01236
Sonnet 5 $0.00039 $0.00495
Haiku 4.5 $0.00019 $0.00247

Measured 3d ago against content hash 09fdcd034134, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

simplicio-autoresearch 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 3d 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.

.claude/skills/simplicio-autoresearch/SKILL.md · 166 lines

How it starts

The opening of the file, as written. The whole thing — 166 lines — stays where its author put it; the contents beside it link to each section on GitHub.

simplicio-autoresearch — evolutionary optimize-by-metric loop

Credit: Andrej Karpathy's autoresearch pattern (https://github.com/balukosuri/Andrej-Karpathy-s-Autoresearch-As-a-Universal-Skill) and the ECC bundle's autoresearch-agent, adapted into a first-class simplicio skill. The upstream loop is NOT installed raw — it conflicts with this ecosystem's rules (unbounded iteration, no git isolation discipline, single-metric hill-climbing). This skill fixes all three. The mechanical bookkeeping (caps, git actions, journal, plateau math, receipt) is a deterministic, model-free worker (scripts/autoresearch.py) — it never mutates the target itself. You (the LLM driving this skill) propose the mutation; the worker enforces the contract around it.

yool guardrails (§11 — MANDATORY, not optional)

Register this skill's runtime as an agent with a hard cap on iterations and budget BEFORE the first mutation — an uncapped loop is a review-blocker, per spec:

### simplicio-autoresearch

- yool_id: `agent.dev.autoresearch`
- authority: dev
- lane: background
- agent_terms:
    cpu_quota_pct: 60
    disk_quota_mb: 100
    timeout_s: 300
    max_iterations: <N>          # MANDATORY — set at `init`, never omitted
    max_token_budget: <N>        # MANDATORY — set at `init`, never omitted

scripts/autoresearch.py init refuses to start (exit 2) without both --max-iterations and --max-token-budget as positive integers — the cap is enforced mechanically, not by convention. record refuses (exit 12) any iteration number beyond the frozen max_iterations.

When to use / when NOT to use

Fit requires: a concrete target file (or small set of files), a command that can score it, and a correctness gate that is genuinely binary.

Good fit (pilots) Why
mapper — TOON encoder heuristics eval = round-trip fixtures (correctness) + measured tokens on real artifacts (score)
dev-cli — prompt template eval = the A/B bench pass-rate, fixed validation set
runtime — local model-ladder tuning eval = ladder success-rate / latency on a fixed task set

Read the full file on GitHub · 166 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. 3d ago First seen · 166 lines · 0 tokens per session scan A 09fdcd034134

Subscribe to this mod's changes

simplicio-autoresearch is a skill published in the GitHub repository wesleysimplicio/simplicio-loop (2 stars, last pushed 4d ago), licensed MIT. It adds 194 tokens to every session and 2,473 once invoked, about $0.0010 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.

Related

Other skills, from other repositories

architecture

Project architecture and file structure conventions for all process types. Use when: (1) Creating new files or modules, (2) Deciding where code should go, (3) Converting single-file components to directories, (4) Reviewing code for structure compliance, (5) Adding new bridges, services, agents, or workers.

iOfficeAI/AionUi · 69 tokens

testing

Testing workflow and quality standards for writing and running tests. Use when: (1) Writing new tests, (2) Adding a new feature that needs tests, (3) Modifying logic that has existing tests, (4) Before claiming a task is complete.

iOfficeAI/AionUi · 55 tokens

strands-review

Local preview of the strands-agents/devtools /strands review agent. Body is the upstream Task Reviewer SOP verbatim — do not paraphrase. Use when the user types /strands-review, asks for a "strands review" of a PR, or wants to anticipate what the remote /strands review GitHub Action will flag. Findings are close but…

strands-agents/harness-sdk · 125 tokens

docs-writer

Draft or rewrite Strands Agents documentation pages. Use when writing new doc pages, rewriting pages that failed audit, drafting sections for existing pages, or writing blog posts and release notes about Strands. Also triggers on "write a doc", "draft a page", "rewrite the quickstart", "add a tutorial for X"…

strands-agents/harness-sdk · 74 tokens

pr-create

Creates a GitHub pull request using the gh CLI. Use when the user asks to create, open, or submit a PR on GitHub.

strands-agents/harness-sdk · 32 tokens

deep-execution

Executes agent-enhanced council queries by spawning parallel Claude subagents that each query a provider, evaluate response quality, ask follow-up questions, and return structured insights with confidence ratings and blind spot analysis. Invoked when the --agents flag is used or when complex architectural decisions…

hex/claude-council · 61 tokens