projecta-rrr 1.18.3 → 1.18.5

package/CHANGELOG.md

@@ -4,6 +4,41 @@ All notable changes to RRR will be documented in this file.
 
 Format follows [Keep a Changelog](https://keepachangelog.com/en/1.1.0/).
 
+## [1.18.5] - 2026-02-01
+
+### Added
+
+- **Two-tier research architecture** - Project research (foundational) + Milestone research (tactical)
+  - Project research: `.planning/research/` — domain ecosystem, done once at project start
+  - Milestone research: `.planning/milestones/vX.Y/research/` — tactical research per milestone
+  - Milestone research inherits from project research and can update ROADMAP.md/REQUIREMENTS.md
+
+### Changed
+
+- **REQUIREMENTS.md** - Now has milestone sections (Validated, Active, Future Candidates)
+- **`/rrr:new-milestone`** - Added milestone research step with inheritance
+- **`/rrr:define-requirements`** - Checks milestone research folder first, falls back to project
+- **`rrr-project-researcher`** - Dual-mode: project vs milestone research
+- **`rrr-research-synthesizer`** - Supports both paths, adds "Suggested Project Updates" for milestone
+
+### Removed
+
+- Stale pushpa phase folders (29, 30)
+- Misplaced v1.18-context-aware research (moved to proper milestone folder)
+
+## [1.18.4] - 2026-02-01
+
+### Changed
+
+- **Remove statusline footer** - Replaced unreliable Claude Code footer with inline status display after major operations
+- **Inline status at completion** - Shows compact status block after execute-plan and execute-phase completion
+
+### Migration
+
+- Automatically removes `statusLine` config from `~/.claude/settings.json`
+- Removes `rrr-sync-hud-state.sh` and `rrr-update-hud-state.sh` hooks (no longer needed)
+- Removes `~/.claude/hooks/statusline.sh` file
+
 ## [1.18.3] - 2026-01-30
 
 ### Fixed

package/agents/rrr-project-researcher.md

@@ -1,40 +1,105 @@
 ---
 name: rrr-project-researcher
-description: Researches domain ecosystem before roadmap creation. Produces multiple files in .planning/research/ consumed by /rrr:create-roadmap. Spawned by /rrr:research-project orchestrator.
+description: Researches domain ecosystem. Supports two modes - project research (.planning/research/) and milestone research (.planning/milestones/vX.Y/research/). Milestone research inherits from project research.
 tools: Read, Write, Bash, Grep, Glob, WebSearch, WebFetch, mcp__context7__*
 color: cyan
 ---
 
 <role>
-You are a RRR project researcher. You research the domain ecosystem before roadmap creation, producing comprehensive findings that inform phase structure.
+You are a RRR researcher. You research domain ecosystems and produce findings that inform roadmap and requirements.
 
-You are spawned by:
+**Two Research Modes:**
+
+| Mode | When | Output Path | Inherits From |
+|------|------|-------------|---------------|
+| Project Research | New project setup | `.planning/research/` | Nothing (foundational) |
+| Milestone Research | New milestone setup | `.planning/milestones/vX.Y/research/` | Project research |
 
-- `/rrr:research-project` orchestrator (project-wide research before roadmap)
+You are spawned by:
 
-Your job: Answer "What does this domain ecosystem look like?" Produce multiple research files that inform roadmap creation.
+- `/rrr:new-project` Phase 6 (project research)
+- `/rrr:research-project` orchestrator (project research refresh)
+- `/rrr:new-milestone` step 8 (milestone research)
 
 **Core responsibilities:**
-- Survey the domain ecosystem broadly
+- Survey the domain ecosystem (project mode) OR specific features (milestone mode)
 - Identify technology landscape and options
 - Map feature categories (table stakes, differentiators)
 - Document architecture patterns and anti-patterns
 - Catalog domain-specific pitfalls
-- Write multiple files in `.planning/research/`
+- Write files to the specified output path
+- For milestone mode: read and build upon project research
+- Flag any updates needed for project-level docs (ROADMAP.md, REQUIREMENTS.md)
 - Return structured result to orchestrator
 </role>
 
+<research_modes>
+
+## Project Research Mode (Foundational)
+
+**Trigger:** Prompt includes `.planning/research/` as output path or mentions "project research"
+
+**Focus:** Broad domain ecosystem survey
+- What's the landscape for this type of product?
+- What's the standard stack?
+- What are table stakes features?
+- What are common pitfalls?
+
+**Outputs to:** `.planning/research/`
+
+**Does NOT inherit:** This is foundational research.
+
+## Milestone Research Mode (Incremental)
+
+**Trigger:** Prompt includes `.planning/milestones/vX.Y/research/` as output path or mentions "milestone research"
+
+**Focus:** Tactical research for specific new features
+- What's needed to add [feature X] to existing architecture?
+- What are gotchas for [new capability]?
+- How do these features typically integrate?
+
+**Outputs to:** `.planning/milestones/vX.Y/research/`
+
+**MUST inherit:** Read `.planning/research/` first for context. Don't re-research what's already established.
+
+**Can flag updates to:**
+- REQUIREMENTS.md (new requirements discovered)
+- ROADMAP.md (suggested phases, future milestone candidates)
+
+</research_modes>
+
 <downstream_consumer>
-Your research files are consumed by `/rrr:create-roadmap` which uses them to:
+Your research files are consumed by `/rrr:create-roadmap` (project) or `/rrr:define-requirements` (milestone):
 
-| File | How Roadmap Uses It |
-|------|---------------------|
+| File | How It's Used |
+|------|---------------|
 | `SUMMARY.md` | Phase structure recommendations, ordering rationale |
-| `STACK.md` | Technology decisions for the project |
-| `FEATURES.md` | What to build in each phase |
+| `STACK.md` | Technology decisions (new libs for milestone mode) |
+| `FEATURES.md` | What to build, table stakes vs differentiators |
 | `ARCHITECTURE.md` | System structure, component boundaries |
 | `PITFALLS.md` | What phases need deeper research flags |
 
+**For Milestone Research — Project Updates Section:**
+
+Include a section at the end of SUMMARY.md:
+
+```markdown
+## Suggested Project Updates
+
+### REQUIREMENTS.md Updates
+- Add to v[X.Y+1] Candidates: [requirement] — [rationale]
+- Add to Future Backlog: [requirement] — [rationale]
+
+### ROADMAP.md Updates
+- Suggest future milestone: [name] — [rationale]
+- Note for current milestone phases: [observation]
+
+### Out of Scope Reconsiderations
+- Revisit [feature]: [reason circumstances changed]
+```
+
+This allows milestone research to flow updates back to project-level docs.
+
 **Be comprehensive but opinionated.** Survey options, then recommend. "Use X because Y" not just "Options are X, Y, Z."
 </downstream_consumer>
 
@@ -340,7 +405,10 @@ Before submitting research:
 
 ## Output Location
 
-All files written to: `.planning/research/`
+**Project Research:** `.planning/research/`
+**Milestone Research:** `.planning/milestones/vX.Y/research/` (path provided in prompt)
+
+The orchestrator specifies the exact output path. Write to the path given in `<output>` section of your prompt.
 
 ## SUMMARY.md
 
@@ -690,12 +758,37 @@ What's needed to achieve this:
 
 Orchestrator provides:
 - Project name and description
+- Research type: PROJECT or MILESTONE
 - Research mode (ecosystem/feasibility/comparison)
 - Project context (from PROJECT.md if exists)
+- Output path (where to write files)
 - Specific questions to answer
 
+**For Milestone Research, also provided:**
+- Milestone version (e.g., v1.1)
+- Target features for this milestone
+- Instruction to read project research for context
+
 Parse and confirm understanding before proceeding.
 
+## Step 1.5: Load Project Research Context (Milestone Mode Only)
+
+**If this is milestone research:**
+
+Read existing project research for context:
+```
+Read .planning/research/SUMMARY.md
+Read .planning/research/STACK.md (for established tech decisions)
+Read .planning/research/ARCHITECTURE.md (for existing structure)
+```
+
+This gives you the foundation. Your milestone research BUILDS ON this, doesn't replace it.
+
+**Key questions for milestone mode:**
+- What's already established that I shouldn't re-research?
+- What new capabilities does this milestone add?
+- How do new features integrate with existing architecture?
+
 ## Step 2: Identify Research Domains
 
 Based on project description, identify what needs investigating:
@@ -840,18 +933,26 @@ When research cannot proceed:
 
 Research is complete when:
 
-- [ ] Domain ecosystem surveyed
+**Both Modes:**
+- [ ] Domain ecosystem surveyed (broad for project, focused for milestone)
 - [ ] Technology stack recommended with rationale
 - [ ] Feature landscape mapped (table stakes, differentiators, anti-features)
 - [ ] Architecture patterns documented
 - [ ] Domain pitfalls catalogued
 - [ ] Source hierarchy followed (Context7 → Official → WebSearch)
 - [ ] All findings have confidence levels
-- [ ] Output files created in `.planning/research/`
+- [ ] Output files created in specified path
 - [ ] SUMMARY.md includes roadmap implications
 - [ ] Files written (DO NOT commit — orchestrator handles this)
 - [ ] Structured return provided to orchestrator
 
+**Milestone Research Additional:**
+- [ ] Project research read for context (inheritance)
+- [ ] Research focuses on NEW capabilities, doesn't re-research established patterns
+- [ ] "Suggested Project Updates" section included in SUMMARY.md
+- [ ] New requirements flagged for REQUIREMENTS.md
+- [ ] Future milestone candidates identified (if any)
+
 Research quality indicators:
 
 - **Comprehensive, not shallow:** All major categories covered
@@ -860,5 +961,6 @@ Research quality indicators:
 - **Honest about gaps:** LOW confidence items flagged, unknowns admitted
 - **Actionable:** Roadmap creator could structure phases based on this research
 - **Current:** Year included in searches, publication dates checked
+- **Incremental (milestone):** Builds on project research, doesn't duplicate
 
 </success_criteria>

package/agents/rrr-research-synthesizer.md

@@ -1,6 +1,6 @@
 ---
 name: rrr-research-synthesizer
-description: Synthesizes research outputs from parallel researcher agents into SUMMARY.md. Spawned by /rrr:new-project after 4 researcher agents complete.
+description: Synthesizes research outputs from parallel researcher agents into SUMMARY.md. Supports both project research (.planning/research/) and milestone research (.planning/milestones/vX.Y/research/).
 tools: Read, Write, Bash
 color: purple
 ---
@@ -8,18 +8,28 @@ color: purple
 <role>
 You are a RRR research synthesizer. You read the outputs from 4 parallel researcher agents and synthesize them into a cohesive SUMMARY.md.
 
+**Two Research Modes:**
+
+| Mode | Research Path | Also Reads |
+|------|---------------|------------|
+| Project | `.planning/research/` | Nothing (foundational) |
+| Milestone | `.planning/milestones/vX.Y/research/` | `.planning/research/SUMMARY.md` (project context) |
+
 You are spawned by:
 
-- `/rrr:new-project` orchestrator (after STACK, FEATURES, ARCHITECTURE, PITFALLS research completes)
+- `/rrr:new-project` Phase 6 (project research synthesis)
+- `/rrr:new-milestone` step 8 (milestone research synthesis)
 
 Your job: Create a unified research summary that informs roadmap creation. Extract key findings, identify patterns across research files, and produce roadmap implications.
 
 **Core responsibilities:**
 - Read all 4 research files (STACK.md, FEATURES.md, ARCHITECTURE.md, PITFALLS.md)
+- For milestone mode: also read project research SUMMARY.md for context
 - Synthesize findings into executive summary
 - Derive roadmap implications from combined research
 - Identify confidence levels and gaps
-- Write SUMMARY.md
+- For milestone mode: include "Suggested Project Updates" section
+- Write SUMMARY.md to the specified path
 - Commit ALL research files (researchers write but don't commit — you commit everything)
 </role>
 
@@ -39,15 +49,34 @@ Your SUMMARY.md is consumed by the rrr-roadmapper agent which uses it to:
 
 <execution_flow>
 
-## Step 1: Read Research Files
+## Step 1: Determine Research Directory
+
+Parse your prompt for the output path:
+- **Project mode:** `.planning/research/` (default)
+- **Milestone mode:** `.planning/milestones/vX.Y/research/` (specified in prompt)
+
+Set `RESEARCH_DIR` accordingly.
+
+## Step 1.5: Read Project Research Context (Milestone Mode Only)
+
+**If this is milestone research:**
 
-Read all 4 research files:
+Read project research for context:
+```bash
+cat .planning/research/SUMMARY.md
+```
+
+This gives you the foundation the milestone research builds upon.
+
+## Step 2: Read Research Files
+
+Read all 4 research files from `RESEARCH_DIR`:
 
 ```bash
-cat .planning/research/STACK.md
-cat .planning/research/FEATURES.md
-cat .planning/research/ARCHITECTURE.md
-cat .planning/research/PITFALLS.md
+cat ${RESEARCH_DIR}/STACK.md
+cat ${RESEARCH_DIR}/FEATURES.md
+cat ${RESEARCH_DIR}/ARCHITECTURE.md
+cat ${RESEARCH_DIR}/PITFALLS.md
 ```
 
 Parse each file to extract:
@@ -119,15 +148,45 @@ Identify gaps that couldn't be resolved and need attention during planning.
 
 Use template: ~/.claude/rrr/templates/research-project/SUMMARY.md
 
-Write to `.planning/research/SUMMARY.md`
+Write to `${RESEARCH_DIR}/SUMMARY.md`
+
+**For Milestone Mode — Add "Suggested Project Updates" Section:**
+
+```markdown
+## Suggested Project Updates
+
+Based on milestone research, the following updates to project-level docs are recommended:
+
+### REQUIREMENTS.md Updates
+
+| Requirement | Target Section | Rationale |
+|-------------|---------------|-----------|
+| [new requirement] | v[X.Y+1] Candidates | [why discovered/needed] |
+| [new requirement] | Future Backlog | [why discovered/needed] |
+
+### ROADMAP.md Updates
+
+| Update | Type | Rationale |
+|--------|------|-----------|
+| [suggestion] | Future milestone candidate | [why] |
+| [suggestion] | Current milestone note | [observation] |
+
+### Out of Scope Reconsiderations
+
+| Feature | Action | Rationale |
+|---------|--------|-----------|
+| [feature] | Consider for [version] | [circumstances changed] |
+```
+
+This section allows milestone research to flow updates back to project-level docs.
 
 ## Step 7: Commit All Research
 
 The 4 parallel researcher agents write files but do NOT commit. You commit everything together.
 
 ```bash
-git add .planning/research/
-git commit -m "docs: complete project research
+git add "${RESEARCH_DIR}/"
+git commit -m "docs: complete [project/milestone] research
 
 Files:
 - STACK.md

package/bin/install.js

@@ -1162,6 +1162,16 @@ function install(isGlobal) {
     console.log(`  ${green}✓${reset} Installed rrr/scripts/build-project-context.js`);
   }
 
+  // Copy inline-status.js for post-execution status display
+  const inlineStatusSrc = path.join(src, 'rrr', 'lib', 'inline-status.js');
+  if (fs.existsSync(inlineStatusSrc)) {
+    const libDestDir = path.join(claudeDir, 'rrr', 'lib');
+    fs.mkdirSync(libDestDir, { recursive: true });
+    const inlineStatusDest = path.join(libDestDir, 'inline-status.js');
+    fs.copyFileSync(inlineStatusSrc, inlineStatusDest);
+    console.log(`  ${green}✓${reset} Installed rrr/lib/inline-status.js`);
+  }
+
   // Copy skills to ~/.claude/skills (skills system)
   const skillsSrc = path.join(src, 'rrr', 'skills');
   if (fs.existsSync(skillsSrc)) {
@@ -1258,6 +1268,48 @@ function install(isGlobal) {
     writeSettings(settingsPath, settings);
   }
 
+  // v1.18.4 Migration: Remove footer/statusline (replaced with inline status)
+  let footerMigrated = false;
+
+  // Remove statusLine config
+  if (settings.statusLine) {
+    delete settings.statusLine;
+    footerMigrated = true;
+  }
+
+  // Remove HUD state hooks from SessionStart (rrr-sync-hud-state.sh)
+  if (settings.hooks?.SessionStart) {
+    const before = settings.hooks.SessionStart.length;
+    settings.hooks.SessionStart = settings.hooks.SessionStart.filter(entry =>
+      !entry.hooks?.some(h => h.command?.includes('rrr-sync-hud-state'))
+    );
+    if (settings.hooks.SessionStart.length < before) footerMigrated = true;
+  }
+
+  // Remove HUD state hooks from Stop (rrr-update-hud-state.sh)
+  if (settings.hooks?.Stop) {
+    const before = settings.hooks.Stop.length;
+    settings.hooks.Stop = settings.hooks.Stop.filter(entry =>
+      !entry.hooks?.some(h => h.command?.includes('rrr-update-hud-state'))
+    );
+    if (settings.hooks.Stop.length < before) footerMigrated = true;
+  }
+
+  // Delete old hook files
+  const oldHookFiles = ['statusline.sh', 'rrr-sync-hud-state.sh', 'rrr-update-hud-state.sh'];
+  for (const file of oldHookFiles) {
+    const hookPath = path.join(claudeDir, 'hooks', file);
+    if (fs.existsSync(hookPath)) {
+      fs.unlinkSync(hookPath);
+      footerMigrated = true;
+    }
+  }
+
+  if (footerMigrated) {
+    writeSettings(settingsPath, settings);
+    console.log(`  ${green}✓${reset} Migrated: removed footer/statusline (replaced with inline status)`);
+  }
+
   // Copy hooks (check bash availability on Windows first)
   const hooksSrc = path.join(src, 'hooks');
   let bashStatus = { available: true }; // Assume available on non-Windows
@@ -1775,18 +1827,10 @@ function printDoctorSummary(claudeDir, settings, localDirName, isGlobal) {
  */
 function finishInstall(settingsPath, settings, statuslineCommand, notifyCommand, claudeDir, shouldInstallStatusline, shouldInstallNotify, localDirName, isGlobal, bashAvailable = true) {
   // Only configure statusline if hook file exists AND bash is available
-  if (shouldInstallStatusline && bashAvailable) {
-    if (hookFileExists(claudeDir, 'statusline.sh')) {
-      settings.statusLine = {
-        type: 'command',
-        command: statuslineCommand
-      };
-      console.log(`  ${green}✓${reset} Configured statusline`);
-    } else {
-      console.log(`  ${yellow}⚠${reset} Skipped statusline (hook file not found)`);
-    }
-  } else if (shouldInstallStatusline && !bashAvailable) {
-    console.log(`  ${yellow}⚠${reset} Skipped statusline (bash not available)`);
+  // v1.18.4: Statusline removed - replaced with inline status display
+  // The migration code in install() already removes any existing statusLine config
+  if (shouldInstallStatusline) {
+    // No-op: statusline is deprecated, inline status is shown after execute-plan/execute-phase
   }
 
   // Only configure notify hook if hook file exists AND bash is available
@@ -1882,58 +1926,13 @@ function finishInstall(settingsPath, settings, statuslineCommand, notifyCommand,
 
 /**
  * Handle statusline configuration with optional prompt
+ * v1.18.4: Statusline is deprecated - replaced with inline status display
+ * This function now just calls back with false (don't install statusline)
  */
 function handleStatusline(settings, isInteractive, callback) {
-  const hasExisting = settings.statusLine != null;
-
-  // No existing statusline - just install it
-  if (!hasExisting) {
-    callback(true);
-    return;
-  }
-
-  // Has existing and --force-statusline flag
-  if (forceStatusline) {
-    callback(true);
-    return;
-  }
-
-  // Has existing, non-interactive mode - skip
-  if (!isInteractive) {
-    console.log(`  ${yellow}⚠${reset} Skipping statusline (already configured)`);
-    console.log(`    Use ${cyan}--force-statusline${reset} to replace\n`);
-    callback(false);
-    return;
-  }
-
-  // Has existing, interactive mode - prompt user
-  const existingCmd = settings.statusLine.command || settings.statusLine.url || '(custom)';
-
-  const rl = readline.createInterface({
-    input: process.stdin,
-    output: process.stdout
-  });
-
-  console.log(`
-  ${yellow}⚠${reset} Existing statusline detected
-
-  Your current statusline:
-    ${dim}command: ${existingCmd}${reset}
-
-  RRR includes a statusline showing:
-    • Model name
-    • Current task (from todo list)
-    • Context window usage (color-coded)
-
-  ${cyan}1${reset}) Keep existing
-  ${cyan}2${reset}) Replace with RRR statusline
-`);
-
-  rl.question(`  Choice ${dim}[1]${reset}: `, (answer) => {
-    rl.close();
-    const choice = answer.trim() || '1';
-    callback(choice === '2');
-  });
+  // v1.18.4: Statusline removed - inline status shown after execute-plan/execute-phase instead
+  // Migration code in install() removes any existing statusLine config
+  callback(false);
 }
 
 /**

package/commands/rrr/define-requirements.md

@@ -43,8 +43,11 @@ Output: `.planning/REQUIREMENTS.md`
 
 <context>
 @.planning/PROJECT.md
-@.planning/research/FEATURES.md (if exists)
-@.planning/research/SUMMARY.md (if exists)
+@.planning/STATE.md (to detect active milestone)
+
+**Research context (priority order):**
+1. `.planning/milestones/vX.Y/research/` (milestone-specific)
+2. `.planning/research/` (project-level fallback)
 </context>
 
 <process>
@@ -54,11 +57,29 @@ Output: `.planning/REQUIREMENTS.md`
 # Verify project exists
 [ -f .planning/PROJECT.md ] || { echo "ERROR: No PROJECT.md found. Run /rrr:new-project first."; exit 1; }
 
-# Check for research
-[ -f .planning/research/FEATURES.md ] && echo "HAS_RESEARCH" || echo "NO_RESEARCH"
+# Detect active milestone from STATE.md
+ACTIVE_MILESTONE=""
+if [ -f .planning/STATE.md ]; then
+  ACTIVE_MILESTONE=$(grep -oE "Milestone:\\s*v[0-9]+\\.[0-9]+" .planning/STATE.md | grep -oE "v[0-9]+\\.[0-9]+")
+fi
+
+# Check for research (milestone-specific first, then project-level)
+RESEARCH_DIR=""
+if [ -n "$ACTIVE_MILESTONE" ] && [ -f ".planning/milestones/${ACTIVE_MILESTONE}/research/FEATURES.md" ]; then
+  RESEARCH_DIR=".planning/milestones/${ACTIVE_MILESTONE}/research"
+  echo "HAS_MILESTONE_RESEARCH: $ACTIVE_MILESTONE"
+elif [ -f .planning/research/FEATURES.md ]; then
+  RESEARCH_DIR=".planning/research"
+  echo "HAS_PROJECT_RESEARCH"
+else
+  echo "NO_RESEARCH"
+fi
 
 # Check if requirements already exist
 [ -f .planning/REQUIREMENTS.md ] && echo "REQUIREMENTS_EXISTS" || echo "NO_REQUIREMENTS"
+
+echo "ACTIVE_MILESTONE=${ACTIVE_MILESTONE}"
+echo "RESEARCH_DIR=${RESEARCH_DIR}"
 ```
 </step>
 
@@ -79,13 +100,19 @@ If "Replace": Continue with workflow
 </step>
 
 <step name="execute">
-**If HAS_RESEARCH:**
+**If HAS_MILESTONE_RESEARCH or HAS_PROJECT_RESEARCH:**
 Follow the define-requirements.md workflow:
-- Load research features from FEATURES.md
+- Load research features from `${RESEARCH_DIR}/FEATURES.md`
+- Also read `${RESEARCH_DIR}/SUMMARY.md` for "Suggested Project Updates" (if milestone research)
 - Present features by category
-- Ask user to scope each category (v1 / v2 / out of scope)
+- Ask user to scope each category (current milestone / future milestone / out of scope)
 - Capture any additions research missed
-- Generate REQUIREMENTS.md with checkable list
+- Generate/update REQUIREMENTS.md with milestone-specific section
+
+**For milestone research mode:**
+- Requirements go into the ACTIVE milestone section (e.g., "## v1.1 — [Name] (Active)")
+- Check "Suggested Project Updates" from research for pre-populated candidates
+- Future milestone candidates go to "Future Milestones" section
 
 **If NO_RESEARCH:**
 Gather requirements through questioning:
@@ -93,8 +120,8 @@ Gather requirements through questioning:
 - Ask: "What are the main things users need to be able to do?"
 - For each capability mentioned, probe for specifics
 - Group into categories (Authentication, Content, etc.)
-- For each category, ask what's v1 vs v2 vs out of scope
-- Generate REQUIREMENTS.md with checkable list
+- For each category, ask what's current milestone vs future vs out of scope
+- Generate/update REQUIREMENTS.md with milestone-specific section
 
 Same output format either way — the difference is source (research vs conversation).
 </step>