القائمة
Archive completed research milestone and prepare for next phase of investigation
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
Add research phase to end of current milestone in roadmap
Capture idea or task as todo from current research conversation context
Prepare a paper for arXiv submission with validation and packaging
Audit research milestone completion against original research goals
Create a hypothesis branch for parallel investigation of an alternative approach
List pending research todos and select one to work on
| name | gpd-complete-milestone |
| description | Archive completed research milestone and prepare for next phase of investigation |
| argument-hint | <version> |
| context_mode | project-required |
| requires | {"files":[".gpd/ROADMAP.md"],"recommended":["audit-milestone"]} |
| allowed-tools | ["read_file","write_file","shell"] |
<codex_runtime_notes> Codex shell compatibility:
gpd on PATH.GPD_ACTIVE_RUNTIME=codex uv run gpd ....
</codex_runtime_notes>Purpose: Create a historical record of completed research (e.g., "analytical derivation complete" or "numerical validation done"), archive milestone artifacts (roadmap + requirements), and prepare for the next stage of investigation. Output: Milestone archived (roadmap + requirements), PROJECT.md evolved, git tagged.
<execution_context> Load these files NOW (before proceeding):
Mark a completed research stage (v1.0, v1.1, v2.0) as done. Creates historical record in MILESTONES.md, performs full PROJECT.md evolution review, reorganizes ROADMAP.md with milestone groupings, archives completed research, and tags the release in git. Prepares handoff notes for the next research stage.
<required_reading>
.gpd/ROADMAP.md.gpd/REQUIREMENTS.md.gpd/PROJECT.md</required_reading>
<archival_behavior>
When a research milestone completes:
.gpd/milestones/v[X.Y]-ROADMAP.md.gpd/milestones/v[X.Y]-REQUIREMENTS.mdContext Efficiency: Archives keep ROADMAP.md constant-size and REQUIREMENTS.md milestone-scoped.
ROADMAP archive uses templates/milestone-archive.md -- includes milestone header (status, phases, date), full phase details, milestone summary (decisions, key findings, open questions).
REQUIREMENTS archive contains all requirements marked complete with outcomes, traceability table with final status, notes on changed requirements.
</archival_behavior>
Use roadmap analyze for comprehensive readiness check:
ROADMAP=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local roadmap analyze)
This returns all phases with plan/summary counts and disk status. Use this to verify:
disk_status == 'complete' for each.progress_percent should be 100%.Present:
Milestone: [Name, e.g., "v1.0 Model Derivation"]
Includes:
- Phase 1: Hamiltonian Setup (2/2 plans complete)
- Phase 2: Mean-Field Solution (2/2 plans complete)
- Phase 3: Numerical Validation (3/3 plans complete)
- Phase 4: Limiting Cases (1/1 plan complete)
Total: {phase_count} phases, {total_plans} plans, all complete
if [ -f .gpd/config.json ]; then cat .gpd/config.json; else echo "WARNING: config.json not found — using defaults"; fi
Auto-approved: Milestone scope verification
[Show breakdown summary without prompting]
Proceeding to stats gathering...
Proceed to gather_stats.
Ready to mark this research milestone as complete?
(yes / wait / adjust scope)
Wait for confirmation.
Calculate milestone statistics:
git log --oneline --grep="feat(" | head -20
git diff --stat FIRST_COMMIT..LAST_COMMIT | tail -1
find . -name "*.py" -o -name "*.tex" -o -name "*.ipynb" | xargs wc -l 2>/dev/null
git log --format="%ai" FIRST_COMMIT | tail -1
git log --format="%ai" LAST_COMMIT | head -1
Present:
Milestone Stats:
- Phases: [X-Y]
- Plans: [Z] total
- Research tasks: [N] total (from phase summaries)
- Files modified: [M]
- Lines of code/text: [LOC] (Python/LaTeX/notebooks)
- Timeline: [Days] days ([Start] -> [End])
- Git range: feat(XX-XX) -> feat(YY-YY)
Extract one-liners from SUMMARY.md files using summary-extract:
# For each phase in milestone, extract one-liner
for summary in .gpd/phases/*-*/*-SUMMARY.md; do
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local summary-extract "$summary" --field one_liner | /Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local json get .one_liner --default ""
done
Extract 4-6 key research accomplishments. Present:
Key accomplishments for this milestone:
1. [Research achievement from phase 1]
2. [Research achievement from phase 2]
3. [Research achievement from phase 3]
4. [Research achievement from phase 4]
5. [Research achievement from phase 5]
Note: MILESTONES.md entry is now created automatically by gpd milestone complete in the archive_milestone step. The entry includes version, date, phase/plan/task counts, and accomplishments extracted from SUMMARY.md files.
If additional details are needed (e.g., researcher-provided "Key Findings" summary, git range, LOC stats), add them manually after the CLI creates the base entry.
Full PROJECT.md evolution review at milestone completion.
Read all phase summaries:
cat .gpd/phases/*-*/*-SUMMARY.md
Full review checklist:
"What This Is" accuracy:
Core Research Question check:
Requirements audit:
Validated section:
- [x] [Requirement] -- v[X.Y]Active section:
Out of Scope audit:
Context update:
Key Decisions audit:
Constraints check:
Update PROJECT.md inline. Update "Last updated" footer:
---
_Last updated: [date] after v[X.Y] milestone_
Example full evolution (v1.0 -> v1.1 prep):
Before:
## What This Is
A systematic study of phase transitions in the 2D Hubbard model using DMFT.
## Core Research Question
What is the nature of the Mott transition at half-filling?
## Requirements
### Validated
(None yet -- complete research to validate)
### Active
- [ ] Derive self-consistency equations for DMFT
- [ ] Implement impurity solver (ED or CTQMC)
- [ ] Map phase diagram (U/t vs T)
- [ ] Compare with exact diagonalization benchmarks
### Out of Scope
- Doped case -- focus on half-filling first
- Cluster extensions (CDMFT) -- single-site first
After v1.0:
## What This Is
A systematic study of the Mott transition in the 2D Hubbard model at half-filling using single-site DMFT with an ED impurity solver.
## Core Research Question
What is the nature of the Mott transition at half-filling?
## Requirements
### Validated
- [x] Derive self-consistency equations for DMFT -- v1.0
- [x] Implement impurity solver (ED) -- v1.0 (6-bath-site solver, converges in <20 iterations)
- [x] Map phase diagram (U/t vs T) -- v1.0 (critical Uc/t = 9.35 +/- 0.15)
### Active
- [ ] Compare with exact diagonalization benchmarks
- [ ] Investigate hysteresis near transition
- [ ] Compute spectral functions at key points
### Out of Scope
- Doped case -- focus on half-filling first
- Cluster extensions (CDMFT) -- single-site captures essential physics at half-filling
- Superconducting order -- not relevant at half-filling
## Context
Completed v1.0 with DMFT self-consistency loop and ED solver.
Phase diagram shows first-order Mott transition with Uc/t = 9.35.
Good agreement with Bulla et al. (1999) NRG results (Uc/t = 9.35).
Minor discrepancy with Georges et al. (1996) at high T -- likely finite bath-size effect.
Step complete when:
Generate a structured research digest that serves as the primary handoff artifact between milestone completion and paper writing. This digest lives at .gpd/milestones/v[X.Y]/RESEARCH-DIGEST.md and captures the full research story in a form optimized for downstream consumption by the paper-writing workflow.
Step 0 -- Ensure digest directory exists:
mkdir -p .gpd/milestones/v[X.Y]
Step 1 -- Run history-digest for foundation data:
DIGEST=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local history-digest)
This returns per-phase summaries with one-liners, provides/requires, and dependency graph data.
Step 2 -- Read source materials:
# All SUMMARY.md files from this milestone's phases
cat .gpd/phases/*-*/*-SUMMARY.md
# Research state
cat .gpd/state.json
# Convention catalog (full reference)
cat .gpd/CONVENTIONS.md
# Convention lock (machine-readable, from state.json)
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local --raw convention list
# Original objectives
cat .gpd/REQUIREMENTS.md
Step 3 -- Compose RESEARCH-DIGEST.md:
Create .gpd/milestones/v[X.Y]/RESEARCH-DIGEST.md with the following structure:
# Research Digest: v[X.Y] [Milestone Name]
Generated: [date]
Milestone: v[X.Y]
Phases: [first]-[last]
## Narrative Arc
[One paragraph summarizing the research story from first phase to last. This is not a
list of what was done -- it is the logical thread that connects the initial question
through intermediate steps to the final result. Written in a way that maps naturally
to a paper's Introduction -> Methods -> Results flow.]
## Key Results
| Phase | Result | Equation / Value | Validity Range | Confidence |
| ----- | ------------- | ------------------------------------------------------- | ------------------ | ------------------- |
| [N] | [Description] | [Equation or numerical value from intermediate_results] | [Parameter regime] | [From verification] |
| ... | ... | ... | ... | ... |
Source: state.json intermediate_results, organized by phase ordering.
## Methods Employed
[Extracted from the `methods` field in SUMMARY.md frontmatter across all phases.
Listed in the order they were introduced, noting which phase introduced each method.]
- **Phase [N]:** [Method name] -- [Brief description]
- **Phase [M]:** [Method name] -- [Brief description]
- ...
## Convention Evolution
[Timeline of convention changes extracted from `gpd --raw convention list`, showing
when each convention was established or modified. This ensures the paper writer uses the
final, settled notation consistently.]
| Date / Phase | Convention | Description | Status |
| ------------ | ----------------- | --------------- | ------------------- |
| Phase [N] | [Symbol/notation] | [What it means] | Active / Superseded |
| ... | ... | ... | ... |
## Figures and Data Registry
[All figures and data files created during this milestone, with paths and descriptions.
Extracted from phase directories and SUMMARY.md artifacts sections.]
| File | Phase | Description | Paper-ready? |
| ------ | ----- | --------------- | ------------ |
| [path] | [N] | [What it shows] | Yes/No |
| ... | ... | ... | ... |
## Open Questions
[From state.json open_questions field. These become the "Future Work" section of a
paper or motivate the next milestone.]
1. [Open question 1]
2. [Open question 2]
3. ...
## Dependency Graph
[The provides/requires graph from history-digest output, rendered as text. Shows how
results flow between phases -- useful for determining which derivation steps are
prerequisites for which results in the paper.]
Phase [N] "[Name]"
provides: [list]
requires: [list]
-> Phase [M] "[Name]"
provides: [list]
requires: [list]
## Mapping to Original Objectives
[Cross-reference results back to REQUIREMENTS.md objectives. Shows which requirements
were fulfilled by which phases/results.]
| Requirement | Status | Fulfilled by | Key Result |
|-------------|--------|-------------|------------|
| [Requirement text] | Complete/Partial | Phase [N] | [Result reference] |
| ... | ... | ... | ... |
Step 4 -- Verify digest completeness:
Before proceeding, confirm:
gpd convention listDelegate archival to gpd CLI:
ARCHIVE=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local milestone complete "v[X.Y]" --name "[Milestone Name]")
The CLI handles:
.gpd/milestones/ directorymilestones/v[X.Y]-ROADMAP.mdmilestones/v[X.Y]-REQUIREMENTS.md with archive headerExtract from result: version, date, phases, plans, tasks, accomplishments, archived.
Verify: Milestone archived to .gpd/milestones/
Note: Phase directories (.gpd/phases/) are NOT deleted -- they accumulate across milestones as raw research history. Phase numbering continues (v1.0 phases 1-4, v1.1 phases 5-8, etc.).
Review project-local insights accumulated during this milestone and promote confirmed patterns to the global pattern library. This is the natural time to consolidate learnings — the research is complete, patterns have been validated across multiple phases, and confidence is highest.
Step 1 — Check for project insights:
ls .gpd/INSIGHTS.md 2>/dev/null
If no INSIGHTS.md exists, skip this step.
Step 2 — Search for matching global patterns:
# Check if pattern library is initialized
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local --raw pattern list 2>/dev/null
Step 3 — Review insights for promotion candidates:
Read .gpd/INSIGHTS.md and identify entries that:
For each candidate, check if a matching pattern already exists:
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local --raw pattern search "keyword from insight" 2>/dev/null
Step 4 — Suggest promotions:
If promotion candidates found, present them:
## Pattern Library: Promote Milestone Insights?
Found {N} confirmed insights that could become global patterns:
1. **{insight title}** — confirmed in phases {X}, {Y}
Domain: {domain} Category: {category}
→ `gpd pattern add --domain {domain} --category {category} --title "{title}"`
2. ...
Promote now? (yes / skip / select specific)
If user approves, run pattern add for each selected insight. If user skips, proceed silently.
If no candidates found: Proceed silently (no output needed).
After milestone complete has archived, reorganize ROADMAP.md with milestone groupings, then delete originals:
Reorganize ROADMAP.md -- group completed milestone phases:
# Roadmap: [Research Project Name]
## Milestones
- **v1.0 Model Derivation** -- Phases 1-4 (completed YYYY-MM-DD)
- **v1.1 Validation** -- Phases 5-6 (in progress)
## Phases
<details>
<summary>v1.0 Model Derivation (Phases 1-4) -- COMPLETED YYYY-MM-DD</summary>
- [x] Phase 1: Hamiltonian Setup (2/2 plans) -- completed YYYY-MM-DD
- [x] Phase 2: Mean-Field Solution (2/2 plans) -- completed YYYY-MM-DD
</details>
Then delete original REQUIREMENTS.md after verifying archive exists:
ROADMAP.md is kept and reorganized (above). REQUIREMENTS.md is deleted -- a fresh one is created for the next milestone.
# Verify archive exists before deleting original
if [ -f ".gpd/milestones/v${VERSION}-REQUIREMENTS.md" ]; then
# Use git rm so the deletion is staged for commit (not just a filesystem rm)
git rm .gpd/REQUIREMENTS.md 2>/dev/null || rm .gpd/REQUIREMENTS.md
else
echo "ERROR: Archive not found at .gpd/milestones/v${VERSION}-REQUIREMENTS.md. Refusing to delete original."
exit 1
fi
Commit the reorganized roadmap and milestone archive:
PRE_CHECK=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local pre-commit-check --files .gpd/milestones/ .gpd/ROADMAP.md 2>&1) || true
echo "$PRE_CHECK"
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local commit \
"chore(milestone-v${VERSION}): reorganize roadmap and retire requirements" \
--files .gpd/milestones/ .gpd/ROADMAP.md
Most STATE.md updates were handled by milestone complete, but verify and update remaining fields:
Project Reference:
## Project Reference
See: .gpd/PROJECT.md (updated [today])
**Core research question:** [Current core question from PROJECT.md]
**Current focus:** [Next milestone or "Planning next research stage"]
Accumulated Context:
Check branching strategy and offer merge options.
Use init milestone-op for context, or load config directly:
INIT=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local init milestone-op)
if [ $? -ne 0 ]; then
echo "ERROR: gpd initialization failed: $INIT"
# STOP — display the error to the user and do not proceed.
fi
Extract branching_strategy, phase_branch_template, milestone_branch_template from init JSON.
If "none": Skip to git_tag.
For "per-phase" strategy:
BRANCH_PREFIX=$(echo "$PHASE_BRANCH_TEMPLATE" | sed 's/{.*//')
PHASE_BRANCHES=$(git branch --list "${BRANCH_PREFIX}*" 2>/dev/null | sed 's/^\*//' | tr -d ' ')
For "per-milestone" strategy:
BRANCH_PREFIX=$(echo "$MILESTONE_BRANCH_TEMPLATE" | sed 's/{.*//')
MILESTONE_BRANCH=$(git branch --list "${BRANCH_PREFIX}*" 2>/dev/null | sed 's/^\*//' | tr -d ' ' | head -1)
If no branches found: Skip to git_tag.
If branches exist:
## Git Branches Detected
Branching strategy: {per-phase/per-milestone}
Branches: {list}
Options:
1. **Merge to main** -- Merge branch(es) to main
2. **Delete without merging** -- Already merged or not needed
3. **Keep branches** -- Leave for manual handling
Platform note: If
ask_useris not available, present these options in plain text and wait for the user's freeform response.
ask_user with options: Squash merge (Recommended), Merge with history, Delete without merging, Keep branches.
Squash merge:
CURRENT_BRANCH=$(git branch --show-current)
git checkout main
if [ "$BRANCHING_STRATEGY" = "per-phase" ]; then
for branch in $PHASE_BRANCHES; do
git merge --squash "$branch"
git commit -m "feat: $branch for v[X.Y]"
done
fi
if [ "$BRANCHING_STRATEGY" = "per-milestone" ]; then
git merge --squash "$MILESTONE_BRANCH"
git commit -m "feat: $MILESTONE_BRANCH for v[X.Y]"
fi
git checkout "$CURRENT_BRANCH"
Merge with history:
CURRENT_BRANCH=$(git branch --show-current)
git checkout main
if [ "$BRANCHING_STRATEGY" = "per-phase" ]; then
for branch in $PHASE_BRANCHES; do
git merge --no-ff "$branch" -m "Merge branch '$branch' for v[X.Y]"
done
fi
if [ "$BRANCHING_STRATEGY" = "per-milestone" ]; then
git merge --no-ff "$MILESTONE_BRANCH" -m "Merge branch '$MILESTONE_BRANCH' for v[X.Y]"
fi
git checkout "$CURRENT_BRANCH"
Delete without merging:
if [ "$BRANCHING_STRATEGY" = "per-phase" ]; then
for branch in $PHASE_BRANCHES; do
git branch -d "$branch" 2>/dev/null || git branch -D "$branch"
done
fi
if [ "$BRANCHING_STRATEGY" = "per-milestone" ]; then
git branch -d "$MILESTONE_BRANCH" 2>/dev/null || git branch -D "$MILESTONE_BRANCH"
fi
Keep branches: Report "Branches preserved for manual handling"
Create git tag:
git tag -a v[X.Y] -m "v[X.Y] [Name]
Completed: [One sentence summary of research achieved]
Key findings:
- [Finding 1]
- [Finding 2]
- [Finding 3]
See .gpd/MILESTONES.md for full details."
Confirm: "Tagged: v[X.Y]"
Ask: "Push tag to remote? (y/n)"
If yes:
git push origin v[X.Y]
Commit milestone completion. Only include files that exist (MILESTONE-AUDIT.md is optional — only present if $gpd-audit-milestone was run beforehand).
Build the file list, conditionally including MILESTONE-AUDIT.md if it exists:
COMMIT_FILES=".gpd/milestones/v[X.Y]-ROADMAP.md .gpd/milestones/v[X.Y]-REQUIREMENTS.md .gpd/milestones/v[X.Y]/RESEARCH-DIGEST.md .gpd/MILESTONES.md .gpd/PROJECT.md .gpd/STATE.md"
if [ -f ".gpd/milestones/v[X.Y]-MILESTONE-AUDIT.md" ]; then
COMMIT_FILES="$COMMIT_FILES .gpd/milestones/v[X.Y]-MILESTONE-AUDIT.md"
fi
PRE_CHECK=$(/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local pre-commit-check --files $COMMIT_FILES 2>&1) || true
echo "$PRE_CHECK"
/Users/charlie/.gpd/venv/bin/python -m gpd.runtime_cli --runtime codex --config-dir ./.codex --install-scope local commit "chore: complete v[X.Y] research milestone" --files $COMMIT_FILES
Confirm: "Committed: chore: complete v[X.Y] research milestone"
Milestone v[X.Y] [Name] complete
Research completed:
- [N] phases ([M] plans, [P] tasks)
- [One sentence of key finding / research outcome]
Archived:
- milestones/v[X.Y]-ROADMAP.md
- milestones/v[X.Y]-REQUIREMENTS.md
- milestones/v[X.Y]/RESEARCH-DIGEST.md
Summary: .gpd/MILESTONES.md
Tag: v[X.Y]
---
## > Next Up
**Start Next Research Stage** -- new questions -> literature review -> requirements -> roadmap
`$gpd-new-milestone`
<sub>`/clear` first -> fresh context window</sub>
---
<milestone_naming>
Version conventions:
Names: Short 1-2 words (v1.0 Model Derivation, v1.1 Validation, v1.2 Parameter Sweep, v2.0 Beyond Mean-Field).
</milestone_naming>
<what_qualifies>
Create milestones for: Complete derivation of key result, validated numerical implementation, systematic parameter study complete, paper draft ready, comparison with literature complete.
Do not create milestones for: Every phase completion (too granular), work in progress, exploratory calculations (unless they produce a definitive result).
Heuristic: "Is this a complete, self-contained research result that could be presented or published?" If yes -> milestone. If no -> keep working.
</what_qualifies>
<success_criteria>
Milestone completion is successful when:
pattern add)</success_criteria>
This template is used by the complete-milestone workflow to create archive files in .gpd/milestones/.
Status: Completed {date} Phases: {phase_start}-{phase_end} Total Plans: {total_plans}
{milestone_description}
{phases_section}
[For each phase in this milestone, include:]
Goal: {phase_goal} Depends on: {depends_on} Plans: {plan_count} plans
Plans:
Details: {phase_details_from_roadmap}
For decimal phases, include (INSERTED) marker:
Goal: Fix divergence in loop integral discovered during calculation Depends on: Phase 2 Plans: 1 plan
Plans:
Details: {phase_details_from_roadmap}
Key Results: {key_results_from_milestone} [Example:]
Key Equations: {central_equations_and_values} [Example:]
Decimal Phases:
Key Decisions: {decisions_from_project_state} [Example:]
Issues Resolved: {issues_resolved_during_milestone} [Example:]
Issues Deferred: {issues_deferred_to_later} [Example:]
Lessons Learned: {lessons_from_research_process} [Example:]
Open Questions and Shortcuts: {shortcuts_needing_future_work} [Example:]
For current project status, see .gpd/ROADMAP.md
How to fill template:
{placeholders} with actual valuesArchive location:
.gpd/milestones/{milestone-name}.md.gpd/milestones/paper-i-submitted.mdAfter archiving:
<details> tag</execution_context>
**Project files:** - `.gpd/ROADMAP.md` - `.gpd/REQUIREMENTS.md` - `.gpd/STATE.md` - `.gpd/PROJECT.md`User input:
If --dry-run flag is present, show what would be archived and deleted (milestone archive files, requirements deletion, roadmap collapse, git tag) without making any changes.
Follow complete-milestone.md workflow:
Check for audit:
.gpd/v{{version}}-MILESTONE-AUDIT.md$gpd-audit-milestone firstgaps_found: recommend $gpd-plan-milestone-gaps firstpassed: proceed to step 1## Pre-flight Check
{If no v{{version}}-MILESTONE-AUDIT.md:}
No milestone audit found. Run `$gpd-audit-milestone` first to verify
research question coverage, derivation completeness, and cross-phase consistency.
{If audit has gaps:}
Milestone audit found gaps (e.g., unchecked limits, missing error bars,
unsupported claims). Run `$gpd-plan-milestone-gaps` to create phases that
close the gaps, or proceed anyway to defer to the next milestone.
{If audit passed:}
Milestone audit passed. Proceeding with completion.
Verify readiness:
Gather stats:
Extract accomplishments:
Archive milestone:
.gpd/milestones/v{{version}}-ROADMAP.mdArchive requirements:
.gpd/milestones/v{{version}}-REQUIREMENTS.md.gpd/REQUIREMENTS.md (fresh one created for next milestone)5b. Promote patterns:
gpd pattern search to check for duplicatesgpd pattern add for promotion candidatesUpdate PROJECT.md:
<details> (if v1.1+)Commit and tag:
Note: MILESTONES.md is CREATED by gpd milestone complete during this workflow. It does not need to exist beforehand.
chore: archive v{{version}} research milestonegit tag -a v{{version}} -m "[milestone summary]"Offer next steps:
$gpd-new-milestone — start next research milestone (questioning -> literature review -> requirements -> roadmap)<success_criteria>
.gpd/milestones/v{{version}}-ROADMAP.md.gpd/milestones/v{{version}}-REQUIREMENTS.md.gpd/REQUIREMENTS.md deleted (fresh for next milestone)<critical_rules>
$gpd-new-milestone which includes requirements definition