Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.usezombie.com/llms.txt

Use this file to discover all available pages before exploring further.

A single trigger — a webhook, a cron fire, a steer message — opens one stage. The agent reasons, calls tools, and the stage closes when it’s done or when it would otherwise overflow the model’s context window. Most zombies finish a stage in seconds to a couple of minutes and never need any of the controls on this page. Read on if you have a long incident that issues dozens of tool calls and want to understand what stops it.

The three things that end a stage

WhatWhen it firesWhat happens
Stage finishedAgent decides it’s doneStage closes cleanly. Diagnosis posted, state checkpointed.
Context fullPrompt approaches the model’s context capAgent ends the stage with a checkpoint, the runtime opens a fresh stage on the same chain, the new stage memory_recalls where it left off.
Budget breachDaily or monthly dollar cap hitStage terminates, budget_breach recorded. Future stages in the next billing window are unaffected.

Continuation chains — when one stage isn’t enough

If the agent can’t finish in one stage (a forty-tool-call incident on a 200K-token model, for example), it ends the current stage with exit_ok: false and a checkpoint_id. The runtime re-enqueues the same event as a continuation; the next stage opens with a fresh prompt and pulls the prior findings via memory_recall(checkpoint_id). A chain caps at 10 continuations. The 11th attempt is refused, the originating event is labelled chunk_chain_escalate_human, and an operator picks it up. The zombie itself stays alive — only that one chain forfeits.

Knobs you can tune

Set these under x-usezombie.context in TRIGGER.md. Defaults work for almost every workload — leave them alone unless you have a specific reason.
KnobDefaultWhen to touch it
tool_window20Raise if your zombie routinely runs 30+ tool calls and you want fewer continuations. Lower if you want it to compact its findings sooner.
memory_checkpoint_every5Cadence at which the agent is prompted to write a memory snapshot during a long stage. Rarely worth changing.
context_cap_tokensmodel-dependentSet explicitly if your model has no published context cap or differs from the default.
If you’ve never seen a chunk_chain_escalate_human failure label, you don’t need to change any of these.

Common questions

My zombie ran for two minutes — is that normal? Yes. Stages can be seconds (a single webhook → Slack post) to a few minutes (a multi-step diagnosis). What matters is that the stage closes cleanly, not how long it took. Will it run forever? No. Three independent ceilings stop it: the stage’s own context cap (the agent voluntarily ends and continues), the continuation cap (10 chunks max), and the budget cap (daily_dollars / monthly_dollars in TRIGGER.md). Do I need to tune anything? Almost certainly not. The defaults are tuned for the flagship platform-ops zombie and translate well to most workloads. Tune your SKILL.md prose first — it’s where the actual reasoning lives.

See also

  • Memory — the API the agent uses to checkpoint findings during a long stage.
  • Budgetsdaily_dollars and monthly_dollars ceilings.
  • Authoring a zombie — the full TRIGGER.md reference.