feat(11-02): add parallelization frontmatter to write_phase_prompt

- Include parallelizable, depends_on, files_exclusive in frontmatter
- Rules for parallelizable field determination
- Parallel-aware context section guidance
- Avoid reflexive SUMMARY references when parallel
This commit is contained in:
Lex Christopherson
2026-01-12 14:14:17 -06:00
parent 082c68960f
commit 31a77ae141

View File

@@ -427,13 +427,33 @@ Use template from `~/.claude/get-shit-done/templates/phase-prompt.md`.
**Multiple plans:** Write separate files ({phase}-01-PLAN.md, {phase}-02-PLAN.md, etc.) **Multiple plans:** Write separate files ({phase}-01-PLAN.md, {phase}-02-PLAN.md, etc.)
Each plan follows template structure with: Each plan follows template structure with:
- Frontmatter (phase, plan, type, domain) - Frontmatter (phase, plan, type, parallelizable, depends_on, files_exclusive, domain)
- Objective (plan-specific goal, purpose, output) - Objective (plan-specific goal, purpose, output)
- Execution context (execute-plan.md, summary template, checkpoints.md if needed) - Execution context (execute-plan.md, summary template, checkpoints.md if needed)
- Context (@references to PROJECT, ROADMAP, STATE, codebase docs, RESEARCH/DISCOVERY/CONTEXT if exist, prior summaries, source files, prior decisions, deferred issues, concerns) - Context (@references to PROJECT, ROADMAP, STATE, codebase docs, RESEARCH/DISCOVERY/CONTEXT if exist, prior summaries, source files, prior decisions, deferred issues, concerns)
- Tasks (XML format with types) - Tasks (XML format with types)
- Verification, Success criteria, Output specification - Verification, Success criteria, Output specification
**Parallelization frontmatter (from parallelization_aware step):**
```yaml
---
phase: XX-name
plan: NN
type: execute
parallelizable: [true if independent, false if has dependencies]
depends_on: [plan IDs from analysis, or empty array]
files_exclusive: [files only this plan modifies]
domain: [optional]
---
```
**Rules for parallelizable field:**
- `true`: No file overlap with sibling plans, no depends_on entries, no SUMMARY references needed
- `false`: Has dependencies, shares files, or requires sequential execution
**If parallelization disabled in config:** Still include fields but set `parallelizable: false` for all plans.
**Context section population from frontmatter analysis:** **Context section population from frontmatter analysis:**
Inject automatically-assembled context package from read_project_history step: Inject automatically-assembled context package from read_project_history step:
@@ -464,7 +484,17 @@ Inject automatically-assembled context package from read_project_history step:
This ensures every PLAN.md gets optimal context automatically assembled via dependency graph, making execution as informed as possible. This ensures every PLAN.md gets optimal context automatically assembled via dependency graph, making execution as informed as possible.
For multi-plan phases: each plan has focused scope, references previous plan summaries (via frontmatter selection), last plan's success criteria includes "Phase X complete". **Context section population (parallel-aware):**
When parallelization enabled:
- Only include SUMMARY references if this plan genuinely needs decisions/outputs from prior plan
- Avoid reflexive "Plan 02 references Plan 01 SUMMARY" patterns
- Each plan should be as self-contained as possible
When parallelization disabled:
- Include SUMMARY references as before (sequential context chain)
For multi-plan phases: each plan has focused scope, references previous plan summaries only when genuinely needed (via frontmatter selection), last plan's success criteria includes "Phase X complete".
</step> </step>
<step name="git_commit"> <step name="git_commit">