enumeray setup
Guide or resume connector setup and a first model candidate trial.
Use when: Prepare a development or managed installation without copying identifiers and paths between commands.
Role and prerequisites: Connector owner or authorized co-admin; a first-model author also needs the current context grant and model authority. Use a matching installed executable and your API origin. Interactive setup guides login; noninteractive setup uses the existing authenticated profile. See guided setup.
Example
Section titled “Example”enumeray setup --directory enumeray-setupenumeray setup --directory enumeray-setup --non-interactive --approve SETUP_PLAN_SHA256Inputs and options
Section titled “Inputs and options”| Input | Meaning and use |
|---|---|
--directory <DIRECTORY> |
Private checkpoint for this journey; reuse it after interruption. |
--mode <MODE> |
Explicit development or managed mode. Development uses one regular OS account; managed operation separates connector and model identities. |
--file <FILE> |
Complete version-1 setup input with host, connector, startup and optional first_model selections. |
--non-interactive |
Never prompt; require complete input and explicit approvals. |
--approve <APPROVE> |
The exact setup plan SHA-256 displayed by the preceding review. Host apply uses its separate host digest. |
--allow-development-execution |
Acknowledge trusted same-user execution and retain the consent in the reviewed runtime configuration. |
--submit-trial |
Explicitly submit/resume the saved first-model candidate trial after reviewing its analysis. |
--confirm-model-preparation |
Confirm site-specific Slurm allocation, batch and runtime preparation before freezing a generated starter. |
--replace-enrollment |
Deliberately issue a replacement one-time enrollment, invalidating earlier unused documents. An uncertain replacement is not automatically repeated. |
--enrollment-file <ENROLLMENT_FILE> |
Privately import an explicitly issued enrollment document for this connector when recovering an uncertain secret response. |
--adopt-existing |
Explicitly adopt matching existing connector/context/grant/model records; conflicting settings remain unchanged. |
Every command also accepts the profile, API, JSON and timeout options.
Choose a subcommand
Section titled “Choose a subcommand”| Command | Purpose |
|---|---|
status |
Inspect saved setup progress and current authenticated readiness. |
apply-host |
Apply the exact reviewed managed host plan locally as an administrator. |
Result and next step
Section titled “Result and next step”Saves a private plan before mutation and resumes the same registration, enrollment, grant, preflight, candidate and analysis identities. Exit 1 also indicates that review or a required host action is pending.
Follow the displayed role-specific action, start the saved connector configuration explicitly, then resume this directory. Activation remains a separate reviewed release command.
Exact syntax, defaults and accepted values
Guide or resume connector setup and a first model candidate trial
Usage: enumeray setup [OPTIONS] [COMMAND]
Commands: status Inspect saved progress and authenticated readiness without requesting checks apply-host Apply reviewed managed changes locally as an administrator; never loads a user profile help Print this message or the help of the given subcommand(s)
Options: --api <API> User API origin override; credentials never cross origins. Host/runner commands use explicit configuration
--directory <DIRECTORY> Private setup checkpoint; defaults to ./enumeray-setup. Reuse it to resume the same operation
--mode <MODE> Execution identity mode; must match the setup input or saved plan
Possible values: - development: Run trusted model code under one regular OS account - managed: Separate connector and model execution identities
--profile <PROFILE> Named local user API/login profile; host/runner commands use explicit configuration
[default: default]
--config <CONFIG> Private user profiles file; defaults to $XDG_CONFIG_HOME/enumeray/profiles.json or ~/.config/enumeray/profiles.json
--file <FILE> Version-1 setup input containing host, connector and first-model selections
--json Emit user API results as JSON (NDJSON for following logs) and runtime errors on stderr; host/runner output follows its own protocol
--non-interactive Require complete input and explicit approvals; never prompt
--approve <APPROVE> Approve the exact saved plan SHA-256 shown by setup
--timeout <TIMEOUT> User API HTTP timeout in seconds; live logs use it for read inactivity. Does not limit execution time
[default: 60]
--allow-development-execution Acknowledge that trusted model code can access this OS user's files and credentials
--submit-trial Explicitly submit/resume the saved first-model candidate trial
--confirm-model-preparation Confirm site-specific Slurm batch/runtime preparation before freezing a generated starter
--replace-enrollment Deliberately issue a new enrollment document when no usable secret was retained
--enrollment-file <ENROLLMENT_FILE> Import an explicitly issued enrollment after recovering an uncertain secret response
--adopt-existing Explicitly adopt matching existing registration, host and model records
-h, --help Print help (see a summary with '-h')Documentation 205dcb3116ea · public sources 8c70356d340f · built .Download documentation identity.