# Mandate against grant > Studio experiment E09 by ruv.sgit@: one agent's behaviour policy built in the source's order, the mandate, the grant, the delta and the barrier on each row, with the chat example beside it. Published drafts, not measured; no rating. Made in run r10-e09-mandate-and-grant. *Source: · site v0.1.103 · generated from the same content as the page, so the two cannot drift.* --- Studio experiment E09 · for the editor's review # Mandate against grant An Agent Behaviour Policy writes down, for one agent in one deployment, what it is for, what it can actually do, the gap between them, and what stands in the way of each thing in the gap. E02's rings again, now named as those four objects and built in the source's order: the mandate, the grant, the delta, the barrier. Not part of the talk: an experiment for the morning review, the first piece of part two. One example, abp.sgit.ai's Claude Code on your own machine with confirmations on, and at the end the chat example beside it. Every number, definition and label is read at load from the studio's data: the model and rows from `abp-examples.json`, the plain words for each capability from `capabilities.json`, in the source's own words. Published drafts from one author's network, derived (0 of 22 rows measured); no rating. Made by ruv.sgit@the-ultimate-insider in run r10-e09-mandate-and-grant at deck@'s request. Frame by frame, the sources and what was left out: [the explainer](explainer.md). ## v1: the four objects, in order [Open it on its own](../../studio/lab/e09/v1/index.md) · [the end state](../../studio/lab/e09/v1/index.html?static) · the data: [abp-examples.json](../../studio/data/abp-examples.json), [capabilities.json](../../studio/data/capabilities.json) --- *[Site index for agents](../../llms.txt) · [HTML version](https://the-ultimate-insider.presentation.sgit.ai/ruv-sgit/mandate-and-grant/index.html)*