Access and Consequential Actions
Understand selected tools, profile grants, permission modes, approvals, receipts, and reconciliation.
Integration access is the intersection of several decisions. A connected credential alone does not grant every agent every provider tool.
Authorization Checks
Immediately before a run starts, Oblive rechecks:
- organization identity;
- enabled profile;
- explicit profile grant;
- active catalog entry;
- integration status;
- credential readiness;
- selected tools or services; and
- the current authorization epoch.
The resulting capability is scoped to one execution. Disabling an integration, changing credentials, or narrowing access invalidates reusable authorization state.
Tool Selection Snapshots
Tool selection is an explicit snapshot, never a wildcard over a provider’s future inventory:
- first connection defaults to Read-only and selects all tools currently eligible for that mode;
- choosing Ask before writes or Autonomous selects all currently eligible read and consequential tools;
- returning to Read-only selects all currently eligible read-only tools;
- an explicit saved tool list overrides those defaults; and
- newly advertised provider tools remain blocked until an owner selects them.
The Tools section can clear its local selection for review, but it cannot save an empty allowlist. Search filters only the displayed rows; Select all eligible tools still uses the complete current eligible inventory. When a connected provider currently has no eligible tools, Oblive preserves the connection as Needs tools so the owner can change mode, reconnect, or wait for a reviewed inventory without re-entering the credential unnecessarily.
Permission Modes
| Mode | Agent behavior |
|---|---|
| Read-only | Only reviewed read operations are available. Unclear operations fail closed. |
| Ask before writes | Reads may proceed; every consequential write becomes a staged approval before provider execution. |
| Autonomous | Available writes may execute without per-action approval, within the integration, task, resource, profile, and organization boundaries. |
Autonomous does not expose unselected tools, widen provider resource scope, override task budgets, or permit Chat to write. Every task write still carries durable intent and idempotency, and uncertain external outcomes still require reconciliation.
Action Lifecycle
- The task proposes a typed action.
- Oblive snapshots policy, authorization, task version, and intent.
- The action executes immediately in Autonomous mode or waits in Ask before writes mode.
- A fenced execution lease prevents competing executors.
- The provider is called at most once.
- A confirmed result stores a receipt.
- An uncertain transport outcome moves to reconciliation.
Never retry an uncertain effect merely because the client did not receive a response.
Engineering branch pushes use the same lifecycle with privately staged Git request bytes. See GitHub and Native Git for token permissions, supported operations, and branch-protection requirements.
Daily Review Threshold
The organization’s daily action limit is a visibility threshold in its configured timezone, not a kill switch. Reaching it opens one non-blocking notification for that local day while actions continue through their normal approval and execution policy. Task-level action budgets remain hard limits. Review recent actions before raising the threshold.