Plan

Boundline 0.82.0 makes planning readiness a runtime gate, not a chat convention.

What plan Does

• evaluates goal quality first
• evaluates plan quality next
• evaluates backlog quality after plan quality
• evaluates planning analysis only after backlog quality is ready
• records plan_quality_state, plan_quality_findings, and plan_quality_assumptions when present
• records backlog_quality_state, backlog_quality_findings, backlog_task_count, backlog_mvp_scope, and backlog_unmapped_items when a Canon backlog packet is expected or available
• records planning_analysis_state, planning_analysis_findings, and planning_analysis_coverage after the runtime checks goal outcomes, validation anchors, execution handoff inputs, backlog sequencing, and any governed evidence already present in the session
• records additive large-codebase context substrate fields such as typed context_pack_entries, omission_findings, repository_map_state, snapshot_cache_state, and patch_safe_edit_attempts when advanced context selection is active
• stops on one phase_request when a planning gate needs missing validation, missing backlog handoff evidence, or another blocking planning input
• keeps execution handoff withheld until the gate clears

What To Read When It Blocks

Use status, next, and inspect to see the same runtime decision from different surfaces. Do not invent execution from chat history, and do not treat planning_analysis_findings as permission to rewrite Canon or the plan outside the runtime repair path.

When the large-codebase substrate reports a blocking omission, downgraded critical artifact, stale tracked snapshot cache, or unsafe oversized full-read request, planning remains blocked until the runtime can repair or narrow the context safely.