13-task-segmentation.md

13 — Task Segmentation

Segment lifecycle, outcome assessment, and segment analytics.

Purpose

A session can contain multiple tasks. MOA tracks each task as a segment so learning is based on discrete outcomes instead of whole-session guesses.

Segments answer:

• What task was being attempted?
• Which tools and skills were used?
• How many turns and tokens did it cost?
• What outcome was assessed?
• What learning evidence should be derived from the outcome?

Data Model

TaskSegment lives in crates/moa-core/src/types/segments.rs; persistent rows live in Postgres task_segments.

Important fields:

• id
• session_id
• storage_partition_id
• user_id
• scope
• tenant_id
• segment_index
• task_summary
• started_at
• ended_at
• outcome
• assessment
• outcome_confidence
• tools_used
• skills_activated
• turn_count
• token_cost
• previous_segment_id

ActiveSegment is the lighter projection stored in session VO state.

Segment Detection

Query rewriting produces QueryRewriteResult:

• retrieval_query
• source
• reason
• is_new_task
• task_summary

When a turn is prepared, SegmentTracker uses the query rewrite metadata and session events to decide whether to:

• keep the current active segment
• create the first segment
• close the previous segment and start a new one

Segment tracking reads only is_new_task and task_summary. Query rewrite metadata does not define a durable session intent taxonomy and does not choose tools for the agent.

The event log records SegmentStarted and SegmentCompleted events.

Segment Counters

During a turn, the orchestrator records:

• tool names used
• skill names activated
• completed turn count
• token cost

The active VO state and task_segments row stay in sync through session store calls.

Segment Assessment

Segment assessment runs after boundaries such as segment completion, idle turns, cancellation, timeout, or deferred continuation evidence. It does not decide whether the live agent loop continues; deterministic turn state, approvals, cancellation, queued messages, and tool events own that control path.

Assessment combines five signal classes:

Signal	Meaning
Tool outcome	Whether tools completed, failed, or produced useful output
Verification	Whether tests/checks/verification commands succeeded
Continuation	Whether the next user message indicates success, rework, abandonment, or a new task
Self-assessment	Whether the agent response claims completion or uncertainty
Structural	Whether turns, cost, and duration are anomalous for the tenant baseline

The assessor outputs:

• resolved
• partial
• unknown
• failed
• abandoned

Assessment phases:

• immediate: when a segment appears idle or completed
• deferred: after a later user message gives continuation evidence
• final: when cancellation or timeout closes the segment

Each assessment updates the segment row and appends segment_assessed to learning_log. After the assessment is persisted, MOA derives an ExperienceRecord from the segment. The segment remains the measurement boundary; the experience record is the learning object used for attribution, candidates, and task-conditioned ranking.

Materialized Views

Segment rows drive learning views:

View	Use
skill_resolution_rates	Ranks skills by tenant-level resolution outcomes
segment_baselines	Provides structural baselines for segment assessment
task_strategy_success_rates	Ranks skills and other strategy components by task fingerprint

Refresh is handled through the session store's materialized-view refresh path.

Compaction Interaction

Segment events are durable boundaries. History compaction can summarize older events, but segment start/completion records remain part of replay and analytics.

Learning Flow

User messages
  -> query rewrite
  -> segment start/continue/complete
  -> tool and skill counters
  -> segment assessment
  -> experience record
  -> attribution records
  -> learning candidates
  -> learning_log after promotion
  -> skill ranking and memory learning

Task segmentation is the measurement layer that makes the rest of MOA's learning pipeline reliable.