feat(context): implement relevance scoring for context file selection #10127

Open
opened 2026-04-17 03:54:53 +00:00 by HAL9000 · 0 comments
Owner

Background

Context file selection currently relies on static path patterns. A relevance scoring system allows the ACMS pipeline to rank candidate context files by their relevance to the current plan and actor, ensuring the most pertinent files are included within token budget constraints.

Acceptance Criteria

  • RelevanceScorer class implemented with configurable scoring factors
  • Scoring considers file recency, access frequency, and semantic proximity
  • Top-N selection by relevance score respects token budget limits
  • Relevance scores exposed in context assembly debug output
  • Test coverage >= 97%

Metadata

  • Commit Message: feat(context): implement relevance scoring for context file selection
  • Branch: feature/v3.6.0/context-relevance-scoring
  • Milestone: v3.6.0
  • Ref: #5172 EPIC: Advanced Context Strategies — Beyond Basic ACMS Pipeline (v3.6.0)

Subtasks

  • Define RelevanceScorer protocol and scoring model
  • Implement recency, frequency, and semantic scoring factors
  • Implement top-N selection with token budget enforcement
  • Integrate scorer into ACMS context assembly pipeline
  • Write unit and integration tests for scorer

Definition of Done

  • Implementation complete and all acceptance criteria met
  • Tests written and passing (coverage >= 97%)
  • PR reviewed and merged
  • Parent epic updated

Automated by CleverAgents Bot
Supervisor: Epic Planning | Agent: epic-planning-pool-supervisor

## Background Context file selection currently relies on static path patterns. A relevance scoring system allows the ACMS pipeline to rank candidate context files by their relevance to the current plan and actor, ensuring the most pertinent files are included within token budget constraints. ## Acceptance Criteria - [ ] `RelevanceScorer` class implemented with configurable scoring factors - [ ] Scoring considers file recency, access frequency, and semantic proximity - [ ] Top-N selection by relevance score respects token budget limits - [ ] Relevance scores exposed in context assembly debug output - [ ] Test coverage >= 97% ## Metadata - **Commit Message**: `feat(context): implement relevance scoring for context file selection` - **Branch**: `feature/v3.6.0/context-relevance-scoring` - **Milestone**: v3.6.0 - **Ref**: #5172 EPIC: Advanced Context Strategies — Beyond Basic ACMS Pipeline (v3.6.0) ## Subtasks - [ ] Define `RelevanceScorer` protocol and scoring model - [ ] Implement recency, frequency, and semantic scoring factors - [ ] Implement top-N selection with token budget enforcement - [ ] Integrate scorer into ACMS context assembly pipeline - [ ] Write unit and integration tests for scorer ## Definition of Done - Implementation complete and all acceptance criteria met - Tests written and passing (coverage >= 97%) - PR reviewed and merged - Parent epic updated --- **Automated by CleverAgents Bot** Supervisor: Epic Planning | Agent: epic-planning-pool-supervisor
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
cleveragents/cleveragents-core#10127
No description provided.