Agent guardrail decision check
Written by Agorean from what the endpoint says about itself
Tells you whether to proceed, slow down, or stop, before your agent takes its next action.
It returns one guardrail decision — proceed, slow down, or stop — meant to be called once before every agent action. The decision is based on cost budget, chain depth, step cap and retry budget, and request volume pressure. A slow-down decision comes with a reason and a wait time. Policy thresholds are configured separately; this endpoint only reports the decision for one action.
WHEN TO USE THIS
When: I need to check whether my agent should take its next action
For example: Call it before the action and read the decision field.
When: I need to know why the engine wants me to slow down
For example: Read the reason and wait time fields on a slow-down decision.
When: I need to guard against runaway cost or retries
For example: Use this, since the decision factors in cost budget and retry budget.
When: I need to know the thresholds behind a decision
For example: Check the separate /policy endpoint for the configured thresholds.
0.001 USDC
Paid to 0x1a9a…33e5
Your agent buys it
npx agorean buy lst_5uetfyqgbuzw
Buy link
https://redlynr.com/run
IS THIS YOURS?
Claim it with one signature.
Sign with the key of the wallet this endpoint pays (0x1a9a…33e5). Claiming cannot be undone.
claimListing("lst_5uetfyqgbuzw", wallet_proof)