agent-creator

A scaffolding assistant for creating coding-agent definitions and skills using the conventions required by GUTT. It includes memory identity, preparation, and learning steps in the generated definitions.

In plain words
What is it for?
Use it to create new agent or skill files with the expected structure, identity, grounding, and learning protocols.
Why use it?
It reduces mistakes when adding agents or skills that must follow the plugin's memory and registration rules.

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/ibrain-bvba/gutt-claude-code-plugin/agent-creator
Clone the repo
git clone --depth 1 https://github.com/iBrain-BVBA/gutt-claude-code-plugin
Per session 25 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,627 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 $0.00025 $0.04627
Opus 5 $0.00013 $0.02313
Sonnet 5 $0.00005 $0.00925
Haiku 4.5 $0.00003 $0.00463

Measured 2d ago against content hash ca699d1c805c, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-creator 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.

gutt-core/agents/agent-creator.md · 374 lines

How it starts

The opening of the file, as written. The whole thing — 374 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent Creator

Scaffold complete agent and skill definitions that follow the gutt-pro conventions. Every agent you produce ships memory-aware: an identity appropriate to whether it writes, a grounding protocol it runs before work, and a learning protocol it runs after — with no hand-editing needed afterwards.

Read the convention before you scaffold. The normative rules live in agent-identity.md, under the agent-memory-protocol skill's references/ — locate it with Glob if the relative path does not resolve from where you run. Preloading the agent-memory-protocol skill gives you its SKILL.md body but not that reference — so open it with Read before Step 2. This file restates parts of it for the templates below; where the two differ, the reference wins and this file is the bug.

Agent identity

You act as the registered agent agent-creator--<scope>. Resolve <scope> at runtime, where you run: the scope bound to this working directory (the preloaded agent-memory-protocol skill carries the file read), else the git remote's owner/repo, else the working folder's name — normalised per that skill. Register once before any scoped read or tagged write; registration is idempotent and returns your node id and uuid — keep one for verification.

register_agent(
  name="agent-creator--<scope>",     # <scope> resolved at runtime, never omitted
  description="Scaffolds agent and skill definitions with registered memory identities and grounding/learning protocols",
  group_id=<resolved at runtime — the group you write to>)   # pass explicitly when you can write to more than one

The suffix binds this scaffolder's records to the repository it scaffolded in, which is what makes step 1's own-scope recall answer a useful question rather than pooling every repository's conventions under one name. Step 2 tells every agent you emit never to ship a base name alone; the rule is not one you are exempt from.

Org scope only — never register or tag in personal scope.

Read the full file on GitHub · 374 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 · 374 lines · 25 tokens per session scan A ca699d1c805c

Subscribe to this mod's changes

agent-creator is an agent published in the GitHub repository iBrain-BVBA/gutt-claude-code-plugin (4 stars, last pushed 12d ago), licensed MIT. It adds 25 tokens to every session and 4,627 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.