ak analytics disable

Stop indexed analytics reads while retaining the private database and authoritative local sources.

Use ak analytics disable when you want analytics-backed views to use their authoritative live collectors without deleting the existing derived index.

Usage

ak analytics disable

The command accepts no positional arguments and has no command-specific flags.

Preview and apply

ak analytics disable --json --no-interactive
ak analytics disable --yes --json --no-interactive

The first invocation returns an unapplied plan and exits 3. The second writes disabled intent to ~/.agentkit/analytics-state.json and leaves ~/.agentkit/analytics/analytics.db in place. AGENTKIT_HOME changes the base.

Disable is allowed when analytics is uninitialized, enabled, or already disabled, so it is safe to repeat. It refuses deleted state; use enable when you want to opt in again.

The operation does not scan source contents, refresh facts, delete generations, or modify Claude, Codex, or AgentKit activity source files. It starts no scheduler and performs no network, provider, auth, or entitlement work.

Confirmation behavior

When prompts are enabled, only y or yes applies. Any other answer or EOF prints the preview, leaves state unchanged, and currently exits 0.

With --json or --no-interactive, no prompt is read. Without --yes, the command prints the preview and exits 3. There is no dry-run or force flag.

Output

Human preview or success is written to stdout; the interactive prompt is on stderr. JSON uses one kind=analytics.plan envelope on stdout:

data.plan: schema_version, plan_id, operation, status, expires_at
data.applied: boolean
data.status: post-operation status when applied

Disable does not include a scan. For an unapplied preview, use data.plan.status; data.status is currently a zero-value object.

Recovery boundary

Disable is reversible without rebuilding: run ak analytics enable --yes to re-enable the retained store and refresh it before success. If storage changed while the confirmation was open, apply fails instead of acting on unreviewed state.

Exit behavior

Exit	Meaning
`0`	Disable applied, or an enabled prompt was declined/ended at EOF.
`1`	Planning, state validation, storage, contention, or output failed; positional arguments also use `1`.
`2`	A flag was invalid.
`3`	A noninteractive preview completed without `--yes`.

Lifecycle/service failures in JSON mode emit a structured error envelope on stderr. Flag-parse, argument-validation, and final output-render failures can fall outside that path.

On macOS, the CLI-wide first-launch guard can also write ~/.agentkit/.first-run-darwin and print a one-time hint to stderr before this command runs.

CLI conventions