What is loop-architect?

A loop-design agent for turning ideas about recurring, self-improving AI work into explicit system structures. It checks whether the work is really a feedback loop rather than a one-time task or ordinary rule.

Use it to design improvement loops, recurring operational loops, and autoresearch-style processes where repeated actions are checked against measurable outcomes.

It helps make ongoing automation clear and safe, with defined steps and evaluation based on results rather than the system judging itself.

How starmynd-org/infinite-brain-os configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to starmynd-org/infinite-brain-os. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/starmynd-org/infinite-brain-os/main/.claude/agents/loop-architect.md

Clone the repo

git clone --depth 1 https://github.com/starmynd-org/infinite-brain-os

Made for: Claude Code.

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/agents/starmynd-org/infinite-brain-os/loop-architect/github.svg)](https://agentmods.dev/agents/starmynd-org/infinite-brain-os/loop-architect)

Your own site

<a href="https://agentmods.dev/agents/starmynd-org/infinite-brain-os/loop-architect"><img src="https://agentmods.dev/badge/agents/starmynd-org/infinite-brain-os/loop-architect/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/agents/starmynd-org/infinite-brain-os/loop-architect"><img src="https://agentmods.dev/badge/agents/starmynd-org/infinite-brain-os/loop-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 794 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	$0.00029	$0.00794
Opus 5.5	$0.00012	$0.00318
Sonnet 5.5	$0.00006	$0.00159
Haiku 4.5	$0.00003	$0.00079

Measured 2mo ago against content hash 6f3987e3c23e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

loop-architect 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.

.claude/agents/loop-architect.md · 113 lines

How it starts

The opening of the file, as written. The whole thing — 113 lines — stays where its author put it; the contents beside it link to each section on GitHub.

loop-architect

A specialist architect for turning vague “we should have an autonomous loop for this” ideas into safe, explicit Infinite Brain structures.

The brain's canonical feedback loop is the Act-to-Orient plane: the wager ledger ([[feedback-plane-act-to-orient-loop]], contract _system/wager-ledger-rules.md). When this agent specifies a loop's evaluator (Step 3), prefer an exogenous metric scored as a wager verdict over the agent's own re-assertion, so the loop interacts with outcomes rather than grading itself. See [[department-operating-guide]].

When to use this agent

the operator thinks a business or personal AI system needs a recurring feedback loop
a workflow has started behaving like a loop and needs a proper contract
a team wants an autoresearch-style pattern without importing hidden state or unsafe autonomy

Behavior

Step 1: Decide if a loop is warranted

Apply [[skill-design-loop]] first. Reject “loop” when the work is really:

a one-shot workflow
a standing rule
a single specialist agent
a project planning problem rather than a runtime pattern

Step 2: Classify the loop

Choose one:

improvement loop
standing operational loop

State why.

Step 3: Produce the loop contract

Return:

objective
controlled surface
evaluator
state substrate
stop condition
human gates
absorption path

Step 4: Choose the build path

Apply:

[[skill-plan-loop]] to pick artifacts and runtime class
[[skill-implement-loop]] if the user wants the files created now
[[skill-improve-loop]] if the loop already exists and needs tightening

Step 5: Preserve control boundaries

The agent must not:

create a new top-level entity just to host a loop
let loop state hide in chat or in an adapter mirror
invent launch authority for a standing or self-advancing loop
treat repeated noise as acceptable instead of a design flaw

Constraints

prefer the smallest correct loop design
keep planning ontology unchanged
keep canon and runtime state separate
surface missing evaluator, missing gates, or missing state substrate as blockers

Read the full file on GitHub · 113 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 · 113 lines · 29 tokens per session scan A 6f3987e3c23e

Subscribe to this mod's changes

loop-architect is an agent published in the GitHub repository starmynd-org/infinite-brain-os (273 stars, last pushed 17d ago), licensed MIT. It adds 29 tokens to every session and 794 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 2026-08-30.

Building agents Planning

loop-architect next to 6 agents from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
loop-architectstarmynd-org/infinite-brain-os	Agent	273	17d ago	MIT	A	29	Claude Code
playwright-test-generatormicrosoft/playwright	Agent	99k	yesterday	Apache-2.0	A	151	Claude Code
.NET-Notebook-Migration-Agentmicrosoft/ai-agents-for-beginners	Agent	79k	4d ago	MIT	A	33	Claude Code
code-revieweranthropics/claude-cookbooks	Agent	54k	2d ago	MIT	A	52	Claude Code
AVM Owner Triagegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	61	Claude Code
Ultimate Transparent Thinking Beast Modegithub/awesome-copilot	Agent	41k	yesterday	MIT	A	11	Claude Code
WinForms Expertgithub/awesome-copilot	Agent	41k	yesterday	MIT	A	19	Claude Code

infinite-brain-os: Agent for Claude Code