{
"schema_version": 1,
"id": "full-auto",
"description": "Full auto: nothing is asked, nothing is sandboxed, and every capability this build ships as stable-but-off is on. `approval_policy` is never, `sandbox_mode` is danger-full-access, `web_search` is live -- the most capable of the four modes -- and `features.memories`, `features.multi_agent_v2` and `features.recommended_plugins` are enabled, which is all three stable features the 0.151.0 registry leaves off. Measured against that artifact: `config.toml parse ok`, 49 enabled and 3 overridden where the product's own default is 46 and 0. This is a setup posture -- keys in this product's own configuration file. It is not an execution profile and it grants no environment: what it changes is what the product asks you and which of its own tools exist, not what anyone is permitted to run. Higher-authority layers can still clamp it: this file is one layer of ten, and managed requirements constrain rather than merge.",
"sources": [
"https://learn.chatgpt.com/docs/config-file/config-reference"
]
}