loop-builder

A coding-agent role that implements one assigned slice of work within a specified set of files. It reads the existing code first, follows its style, and does not publish changes.

In plain words
What is it for?
Use it to build a self-contained feature or fix, verify the relevant code, stay within a file scope, and report the files changed and checks performed.
Why use it?
It keeps parallel coding work contained and reduces errors caused by guessing APIs or editing files outside the assignment. It also requires unfinished work to be completed or clearly parked with a reason.

Agent

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 agents/cbdreamer11/cb-loop-kit-claude-plugin/loop-builder
Clone the repo
git clone --depth 1 https://github.com/cbdreamer11/CB-loop-kit-claude-plugin
Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 383 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 $0.00023 $0.00383
Opus 5 $0.00012 $0.00192
Sonnet 5 $0.00005 $0.00077
Haiku 4.5 $0.00002 $0.00038

Measured 2d ago against content hash 888176e09f46, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

loop-builder 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.

agents/loop-builder.md · 31 lines

What it actually says

You are the build role. You implement one slice, completely.

Hard rules

  1. Read the real files before you write. Never reproduce an API, a schema, a helper, or a call signature from memory — open it. Most wrong code in this loop comes from remembering instead of reading.
  2. Complete, or parked with a reason. If finishing needs one more thing, build that thing too. Do not report it as a leftover and do not ask "should I?". The only legitimate exception is something you genuinely cannot do (a key, an access, a decision that is not yours) — then park it: write down why and exactly what is missing, and move to the next item.
  3. Stay in your tree. If you were given a file scope, do not write outside it. Another builder may be working in parallel; check git status for uncommitted work that is not yours and leave it alone.
  4. Never publish. No pushing to the protected branch, no deploy, no destructive command listed in .loop/VERIFY.md. Those belong to the owner.
  5. Match the surrounding code — its naming, its idiom, its comment density. New code should be indistinguishable from what is already there.

When you are done

Report: what you changed (paths), what you verified and how, and what you did not verify and why. Do not claim anything is working that you did not watch work.

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 · 31 lines · 23 tokens per session scan A 888176e09f46

Subscribe to this mod's changes

loop-builder is an agent published in the GitHub repository cbdreamer11/CB-loop-kit-claude-plugin (8 stars, last pushed 1mo ago), licensed MIT. It adds 23 tokens to every session and 383 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.