agent-scope

A setting for choosing the scope of agent identities: project, team, or individual. The scope is the label added to an agent's registered name.

In plain words
What is it for?
Use it to view or set the current agent scope for this working directory.
Why use it?
It gives agents a consistent naming scope so memory registrations are associated with the intended working context.

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/ibrain-bvba/gutt-claude-code-plugin/agent-scope
Clone the repo
git clone --depth 1 https://github.com/iBrain-BVBA/gutt-claude-code-plugin
Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,034 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 $0.00059 $0.01034
Opus 5 $0.00030 $0.00517
Sonnet 5 $0.00012 $0.00207
Haiku 4.5 $0.00006 $0.00103

Measured 2d ago against content hash 8ef44737e3b6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

agent-scope 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/commands/agent-scope.md · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

gutt agent scope

The read or change for this invocation has already been applied, deterministically, by the plugin's UserPromptSubmit hook. It reports the outcome as context alongside this message.

Read the outcome and take exactly one of the three branches below, in this order.

1. No outcome accompanies this message

Then whether anything was applied is unknown — not "nothing happened". The write and the report are separate steps, so a missing report does not mean a missing write. Say that, tell the user to run /gutt-pro:agent-scope show before retrying, and stop. Do not re-run the command and do not guess.

2. The outcome contains the phrase no argument given

Only then, help the user choose one. Check this before branch 3 — an outcome with no argument also reports what is in force, so testing for that first would swallow this case.

  1. Find the labels already in use, so a team converges on one instead of each person inventing their own. Search the graph for existing agent identities and collect the part after -- in the names you find. Read-only — register nothing, write nothing. Search registered names; if what comes back is node IDs, the -- has been collapsed to a single - and the label cannot be split out reliably — say the survey was inconclusive rather than guessing where a name ends.
  2. Check what the candidate would collide with. A registered name is permanent: registration merges on name and group, so reusing a label that belongs to another context silently joins that context's memory to this one, and it cannot be undone or separated afterwards. If a label you are about to suggest already has agents whose work is plainly someone else's, say so and steer away from it.
  3. Ask, using AskUserQuestion — the scope type and the name. Offer the labels from step 1 as options where any exist, and say what each already appears to cover. A team label is the one to prefer when several directories are genuinely one product; a project label when this one should keep its own memory.
  4. Tell the user the exact line to type. You cannot apply it yourself — this command's effect comes from the hook that sees the typed prompt, and a command you invoke never reaches that hook. End with the line, filling in the type and name that were actually chosen, and nothing after it:

Read the full file on GitHub · 82 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 · 82 lines · 59 tokens per session scan A 8ef44737e3b6

Subscribe to this mod's changes

agent-scope is a command published in the GitHub repository iBrain-BVBA/gutt-claude-code-plugin (4 stars, last pushed 12d ago), licensed MIT. It adds 59 tokens to every session and 1,034 once invoked, about $0.0003 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.