msd-core/gsd-core/references/execute-phase-context-guard.md at 4e473c884763ae1714286a24f86d7c5cbe13e50e

Files

Tom Boucher b63a500fad fix(#1505 ): extract context_guard step to reference file; fix allow-test-rule see ref

- Extract execute-phase.md context_guard step prose to
  gsd-core/references/execute-phase-context-guard.md (@-ref lazy load),
  bringing execute-phase.md back under the ADR-857 phase-6 size ceiling
  (92914 < 93166 bytes)
- Fix allow-test-rule comment: add `see #1452` per ADR-456 lint rule
- Update feat-1452 tests to check the reference file for extracted content
- Register execute-phase-context-guard.md in INVENTORY-MANIFEST.json and
  INVENTORY.md Workflow References section
- Regenerate workflow-size-baseline.json after file shrinkage

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

2026-06-20 18:38:20 -04:00

1.1 KiB

Raw Blame History

Context exhaustion guard — context_guard (BEFORE spawning, #1452):

Before spawning any agents for this wave, self-assess context pressure using the degradation signals in references/context-budget.md. Signs of POOR tier (70%+): increasing vagueness, skipped steps, silent partial completion.

Read workflow.context_guard_mode from .planning/config.json (default warn).

Tier	`warn` (default)	`auto`	`off`
PEAK / GOOD	No output	No output	No output
DEGRADING (50-70%)	Emit: "⚠ Context pressure DEGRADING — switching to frontmatter-only reads for remaining waves." Continue.	Same as warn	Skip
POOR (70%+)	Emit: "🛑 Context pressure POOR — risk of context exhaustion. Run `/gsd:pause-work` to checkpoint before this wave, then resume in a fresh session." Continue (user decides).	Invoke `/gsd:pause-work` immediately and halt. Do NOT spawn wave agents.	Skip

The guard is heuristic — no programmatic context-percentage API exists. Use your assessment of degradation signals, not a fixed token count.

1.1 KiB Raw Blame History

1.1 KiB

Raw Blame History