Sculptured docs

Limits and usage

Token limits, agent counts, usage holds and timeouts, with their defaults and enforcement behavior.

Use token limits and output budgets to control usage, and team limits to bound the number and depth of managed subagents. Subagent counts have no default cap; set one if your team can create helpers. Token enforcement depends on CLI usage reports, so an agent can exceed a limit by up to one turn.

flowchart LR
  T["Turn ends"] -- "CLI reports tokens" --> U["Usage so far"]
  U -- "checked every second" --> L{"Over a limit?"}
  L -- "Token limit" --> S["Soft stop: wrap-up, then stop"]
  L -- "Output budget" --> H["Hard stop: agent and its subagents"]
  P["Claude / ChatGPT plan at 90%"] -- "holds" --> N["New work: start a team, spawn an agent"]

Limits you set

LimitWhereWhat it countsWhen it is reached
Token limitAn agent's settings on the Design boardOne of: all tokens, new tokens (all but cache reads), or output tokens, since the run started. By default its subagents count too.The agent is soft-stopped: it is asked for a wrap-up, then stopped.
Output budgetA team's budget, set while the team is stoppedEach agent's output tokens since the run started, from 1 to 1,000,000 per agent, subagents included.The agent and its subagents are hard-stopped. It cannot launch again until a new run starts.
Check timeoutA check command on a workflow stageWall-clock time of the command. Default 600 seconds.The check fails.

Run budgets, and each agent's turn and working-time limits, are in Run budgets.

Protections

On by default; change them in the team's Settings → More settings → Protections (details).

LimitDefault
A whole read of a file is refused above100 KB (images, PDFs and notebooks are never refused)
Shell output an agent sees8,000 characters
Context before an agent writes its notes and continues fresh60,000 tokens, and only once its conversation has doubled from where it started

What agents know

LimitValue
Files a project map looks at20,000 (a larger folder is mapped from its first files)
Areas in the full map / in an agent's briefing80 / 25
Time a start waits for each folder's map20 seconds; then agents start without it
Lessons in an agent's briefingThe 8 newest
One lesson400 characters
Matches one history search returns8
Lessons kept with a project that a briefing lists30
A skill's name / description / instructions64 characters / 1,024 characters / 100 KB

Soft and hard stops

StopWhat happens
Soft stopSculptured interrupts the current turn if there is one, then asks the agent for a summary "within approximately 5,000 additional output tokens". The agent stops when that wrap-up turn ends, after about 5,000 more output tokens, or after 5 minutes, whichever comes first.
Hard stopThe process is stopped with no further turn, along with every subagent it started.

Usage reports can arrive late, so a soft stop can overshoot its 5,000-token estimate. See Control live agents.

Subagents

An agent can start tracked helpers with the spawn_agent tool (agent tools). Sculptured sets no limits on them. A team can set its own in its Settings, under More settings; blank means no limit:

Team settingRefusal the agent gets
Most agents, children included (the team's own agents plus every child)"Team limit you set: 20 agents, children included"
Most levels of children (1 = children cannot start their own)"Team limit you set: 3 levels of children"

The parent's card on the Run board shows the refusal. The settings are saved in the team file as limits.max_agents and limits.max_child_levels.

A spawn is also refused when the parent's budget is used up, the parent is stopping, the team is not running, or the plan is held (below). Claude Code's and Codex's own built-in delegation is turned off, so every helper is an agent you can see on the board.

Plan usage holds

Claude Code and Codex report how much of your Claude or ChatGPT plan is used. When the fullest window that has not reset yet is at 90% or more, Sculptured holds new work:

These readings are for your whole account, so they include use outside Sculptured. Claude Code reports usage only after a Claude agent has run, so before that there is no reading and no hold.

When a turn hits the provider's own limit, Sculptured shows the CLI's message and the reset time if the CLI gave one. Nothing retries it for you; send "continue" after the reset.

Questions and messages

LimitValue
Time a question waits for your answer10 minutes by default. Afterwards the agent decides for itself and says what it decided.
Question text4,096 bytes; longer text is cut and marked.
Answer options in one question20, each label up to 200 bytes
A result handed to the next agent8 KB; longer results are cut and marked "…(truncated)".
One agent report50 findings, each up to 1,024 bytes
Captured text, tool output and errors64 KB per item
Decision log entry4,096 bytes

Starting and stopping

LimitValue
Time for an agent to be ready after launch60 seconds; otherwise it needs attention.
Grace before a stopped agent's process is ended5 seconds
Workflow items' own copies of a repository made at once, per team2

Connected computers

LimitValue
One-time connect key from the Computers pageWorks once, for 30 minutes
Pairing code (when a computer connects without a key)10 minutes
Pairing attempts10 code lookups or approvals per user in 10 minutes; 20 pairing starts per address in 10 minutes
Runner keep-aliveA ping every 25 seconds. The link is dropped after 75 seconds with nothing back, and the runner reconnects on its own.
Reconnect backoffFrom 0.5 to 5 seconds between tries
How long an action waits for a computer that just dropped off45 seconds, then it is reported as not connected
How long an agent's tool answer is kept for a computer that dropped off6 hours

See When a computer drops off for what these mean in practice.

Hosted sign-in sessions

LimitValue
Session ends after no use30 days (each use renews it)

Usage history

The usage chart for a team keeps up to 240 points per agent; older points are thinned, not lost from the totals. Costs are estimates at API prices and name their source. Claude or ChatGPT plan subscribers are not billed those amounts. See Context and communication.

Next steps