Pipeline Plan 211

Plan complete and saved to docs/plans/2026-03-09-memory-pattern-effectiveness-tracker.md.

Summary

Key discovery: The lib/memory-effectiveness.sh module (507 lines) already exists with core injection/outcome tracking, scoring, ranking, pruning, and reporting. The main work is enhancement + integration, not building from scratch.

What exists (reusing):

• memeff_track_injection(), memeff_track_outcome(), memeff_score_pattern(), memeff_rank_patterns(), memeff_prune_ineffective(), memeff_proactive_score(), memeff_report()
• 16 passing tests in sw-memory-effectiveness-test.sh

What the plan adds (7 tasks):

1. memeff_score_issue() — Pre-pipeline issue-to-pattern scoring (keyword overlap + label match + file overlap + effectiveness weighting, 0-100)
2. Enhanced outcome tracking — Adds pattern_injected, failure_occurred, failure_type_matched, failure_prevented fields
3. Enhanced reporting — Injection success rate, strategic export format, patterns-needing-refinement list
4. Hook wiring — Connects memeff_on_injection in sw-loop.sh and memeff_on_pipeline_complete in pipeline-commands.sh + daemon-dispatch.sh
5. CLI subcommands — shipwright memory effectiveness [text|json|strategic] and shipwright memory score-issue
6. Integration tests — E2E flow tests for success/failure paths
7. Full test suite validation — Regression check across pipeline, memory, and daemon tests

Files touched: 5 (modify only, no new files)

Two execution options:

1. Subagent-Driven (this session) — I dispatch fresh subagent per task, review between tasks, fast iteration

2. Parallel Session (separate) — Open new session with executing-plans, batch execution with checkpoints

Which approach?