An objective represents a transformation goal a customer is driving with Klarity Advisor — for example, “shorten close from 7 to 3 days” or “find duplicate invoice processing across geographies”. Each objective accumulates findings, actions, activity, and notes over time. Use objective tools when the user is continuing in-flight transformation work, not starting from scratch. The objective is the persistent record of prior thinking that the Advisor (or your agent) can pick up from.Documentation Index
Fetch the complete documentation index at: https://developers.klarity.ai/llms.txt
Use this file to discover all available pages before exploring further.
What an objective holds
| Object | What it represents |
|---|---|
| Findings | Discrete observations or insights the Advisor surfaced |
| Actions | Concrete next steps proposed against findings |
| Notes | Agent notebook entries, user-supplied context, or run summaries |
| Activity | Append-only log of what the Advisor (and the user) did |
| Agent state | Persisted checkpoint of the Advisor’s reasoning loop |
When to load an objective
- The user names an ongoing initiative (“our close-cycle work”, “the AP duplication project”).
- You’re about to spin up a fresh analysis on a process or value stream — check first whether Advisor has already worked on it.
- You need to resume a prior run without losing context.
| User question | Tool to start with |
|---|---|
| ”What’s the state of our [objective]?” | get_objective_context_bundle |
| ”What findings has Advisor surfaced?” | list_objective_findings |
| ”What actions are proposed?” | list_objective_actions |
| ”What did the user steer on?” | list_objective_items_with_active_steering |
| ”Resume from the last run” | get_objective_agent_state |
The single best entry point
get_objective_context_bundle returns the stitched bundle — objective + findings + actions + activity + notes + agent state — in one call. Use it for orientation before drilling into individual items.
See Tools by purpose → Objectives for the full toolset.