moveros 4.0.8 → 4.1.0

package/src/workflows/overview.md

@@ -1,203 +1,203 @@
----
-description: Project context refresh. Shows current state, processes notes, flags stale projects. Absorbs /project-notes.
----
-
-# Project Overview (V4 - NEW)
-
-**Trigger:** User runs `/overview` or `/overview [project name]`
-**Role:** Project Intelligence Officer.
-**Task:** Refresh project context. Surface state, process notes, flag issues. Like Claude's `/init` for projects.
-**Risk Tier:** LOW
-
----
-
-## 0. PRE-FLIGHT
-
-Global Rules are pre-loaded as CLAUDE.md / GEMINI.md. Output `[Rules Loaded - DATE]`.
-
-LOAD (Light — Active_Context is the cache):
-1. Active_Context.md (has Strategic Verdict S1, Technical Context S4, Active Sessions S6)
-2. Active project's plan.md, project_state.md, project_brief.md (Step 2 reads)
-3. Last 2-3 Daily Notes (P5) — provides recent session context. When returning to a project after days away, these show what the user was actually doing, what was completed, and where they left off. Read via subagent (model: haiku) to keep main context clean.
-NOTE: Strategy context comes from Active_Context S1. If Active_Context is stale (>2 days), flag it and suggest /analyse-day first.
-
-CONSTRAINTS:
-- All claims grounded in project files
-- Confidence < 3 = flag, don't assert
-- Append-only: never delete notes or history
-- Windows: no && in commands
-
----
-
-## 1. PROJECT IDENTIFICATION
-
-If project specified: use that project.
-If not specified: ask "Which project? (Or type 'all' for a portfolio overview)"
-
-Locate the project folder in `01_Projects/`.
-
----
-
-## 2. PROJECT SCAN
-
-Read these files (if they exist):
-
-1. **project_state.md** - Knowledge journal, notes, solutions ledger
-2. **plan.md** - Task roadmap and execution log
-3. **project_brief.md** - Original project brief
-4. **File tree** - List all files in the project folder
-
-Extract:
-- **Technology Stack:** From brief or plan.md
-- **Current Phase:** From plan.md (what's checked vs unchecked)
-- **Progress:** [X] of [Y] tasks complete ([Z]%)
-- **Last Activity:** Date of last execution log entry
-- **Open Blockers:** From Active_Context.md or plan.md
-- **Key Decisions:** From changelog in project_state.md
-- **Recent Context:** From last 2-3 Daily Notes — what sessions touched this project, what was accomplished, where the user left off
-
----
-
-## 3. STALENESS CHECK
-
-Calculate days since last activity:
-
-| Days Since Activity | Status | Action |
-|---------------------|--------|--------|
-| 0-7 | Active | Normal overview |
-| 8-14 | Cooling | "This project hasn't been touched in [X] days." |
-| 15-30 | Cold | "Cold Project warning. Consider: resume, defer, or archive?" |
-| 30+ | Frozen | "FROZEN PROJECT ([X] days). Archive to `04_Archives/` or reboot?" |
-
----
-
-## 4. PROCESS NOTES (Absorbed from /project-notes)
-
-**Target:** `project_state.md` > `# Notes`
-
-1. Read all text under `# Notes`.
-2. Identify: concrete action items, bug reports, feature requests.
-3. **Show user proposed additions before writing:** "I found [X] action items in your Notes. Proposed additions to plan.md: [list]. Approve?"
-4. **STOP. Wait for user confirmation.**
-5. After approval: Add to appropriate Phase in `plan.md`.
-6. For architectural changes: Update `## ARCHITECTURE` in `plan.md`.
-7. Mark processed notes: Append `[Processed YYYY-MM-DD]` next to handled notes.
-8. Do NOT delete original text.
-
----
-
-## 5. OUTPUT THE OVERVIEW
-
-Present a structured summary:
-
-```markdown
-## Project: [Name]
-
-**Status:** [Active/Cooling/Cold/Frozen]
-**Phase:** [Current phase from plan.md]
-**Progress:** [X/Y] tasks ([Z]%)
-**Last Activity:** [Date]
-**Stack:** [Technologies]
-
-### Open Tasks (Top 5)
-1. [ ] [Task] - [Phase]
-2. [ ] [Task] - [Phase]
-
-### Blockers
-- [Blocker 1]
-- [Blocker 2] (or "None")
-
-### Recent Decisions
-- [Decision from changelog]
-
-### Notes Processed
-- [X] notes converted to tasks
-- [Y] notes already processed
-```
-
----
-
-## 5.5. ACTIVE_CONTEXT SYNC
-
-After presenting the overview, update `Active_Context.md` Section 4 (Technical Context):
-- **Active Project:** [Project name] — [Path]
-- **Tech Stack:** [From brief/plan]
-- **Current Bottleneck:** [Top blocker, or "Execution" if none]
-
-This keeps the state bus current for the next session.
-
----
-
-## 6. PORTFOLIO MODE (if 'all' requested)
-
-For each project in `01_Projects/`:
-- Read plan.md (if exists). If no plan.md: Status = "Unplanned", Progress = "N/A", Next Action = "Run /ignite or /init-plan"
-- Calculate status and progress
-- Output a summary table:
-
-| Project | Status | Progress | Last Activity | Next Action |
-|---------|--------|----------|---------------|-------------|
-| [Name] | Active | 60% | 2 days ago | [Task] |
-
-Flag any Cold or Frozen projects.
-
----
-
-## 7. SELF-VERIFICATION (CISC)
-
-Before outputting:
-1. All progress percentages match actual task counts?
-2. Staleness calculation correct?
-3. Notes processed correctly (no data loss)?
-
-If any check fails: flag it.
-
----
-
-## 8. SELF-IMPROVEMENT
-
-If user corrected AI during this workflow:
-1. Acknowledge: "I made a mistake."
-2. Identify the rule violated.
-3. Append to Auto_Learnings.md:
-   `* [YYYY-MM-DD] CORRECTION: [Description]. Source: User correction. Confidence: 3.`
-4. If same correction seen 3x+, escalate confidence to 5.
-
----
-
-## 9. PROACTIVE SUGGESTIONS
-
-Based on overview:
-1. If project Cold: "Consider archiving [Project] or scheduling a reboot session." - Confidence: [X]
-2. If high blocker count: "This project has [X] blockers. Prioritize unblocking." - Confidence: [X]
-3. If notes unprocessed: "[X] notes in project_state.md haven't been processed." - Confidence: [X]
-
-"Want me to address any of these?"
-
----
-
-## 10. ANSWER GATING
-
-Do not suggest handoff until:
-- Project overview has been presented to user
-- Staleness status has been acknowledged
-- Any notes processed are confirmed added to plan.md
-
-If pending: "You haven't confirmed [X]. Wait or proceed?"
-
----
-
-## HANDOFF
-
-Sources used: [List files read]
-Confidence range: [X-Y across all outputs]
-
-**Next:**
-- If Active project: "Continue working, or run `/log` when you finish a session."
-- If Cold project: "Run `/reboot` to break through, or archive to `04_Archives/`."
-- If notes processed: "New tasks added to plan.md. Start executing."
-
-Key rules still active:
-- All claims must be grounded
-- Confidence < 3 = ask, don't assert
-- Append-only: never delete history
+---
+description: Project context refresh. Shows current state, processes notes, flags stale projects. Absorbs /project-notes.
+---
+
+# Project Overview (V4 - NEW)
+
+**Trigger:** User runs `/overview` or `/overview [project name]`
+**Role:** Project Intelligence Officer.
+**Task:** Refresh project context. Surface state, process notes, flag issues. Like Claude's `/init` for projects.
+**Risk Tier:** LOW
+
+---
+
+## 0. PRE-FLIGHT
+
+Global Rules are pre-loaded as CLAUDE.md / GEMINI.md. Output `[Rules Loaded - DATE]`.
+
+LOAD (Light — Active_Context is the cache):
+1. Active_Context.md (has Strategic Verdict S1, Technical Context S4, Active Sessions S6)
+2. Active project's plan.md, project_state.md, project_brief.md (Step 2 reads)
+3. Last 2-3 Daily Notes (P5) — provides recent session context. When returning to a project after days away, these show what the user was actually doing, what was completed, and where they left off. Read via subagent (model: haiku) to keep main context clean.
+NOTE: Strategy context comes from Active_Context S1. If Active_Context is stale (>2 days), flag it and suggest /analyse-day first.
+
+CONSTRAINTS:
+- All claims grounded in project files
+- Confidence < 3 = flag, don't assert
+- Append-only: never delete notes or history
+- Windows: no && in commands
+
+---
+
+## 1. PROJECT IDENTIFICATION
+
+If project specified: use that project.
+If not specified: ask "Which project? (Or type 'all' for a portfolio overview)"
+
+Locate the project folder in `01_Projects/`.
+
+---
+
+## 2. PROJECT SCAN
+
+Read these files (if they exist):
+
+1. **project_state.md** - Knowledge journal, notes, solutions ledger
+2. **plan.md** - Task roadmap and execution log
+3. **project_brief.md** - Original project brief
+4. **File tree** - List all files in the project folder
+
+Extract:
+- **Technology Stack:** From brief or plan.md
+- **Current Phase:** From plan.md (what's checked vs unchecked)
+- **Progress:** [X] of [Y] tasks complete ([Z]%)
+- **Last Activity:** Date of last execution log entry
+- **Open Blockers:** From Active_Context.md or plan.md
+- **Key Decisions:** From changelog in project_state.md
+- **Recent Context:** From last 2-3 Daily Notes — what sessions touched this project, what was accomplished, where the user left off
+
+---
+
+## 3. STALENESS CHECK
+
+Calculate days since last activity:
+
+| Days Since Activity | Status | Action |
+|---------------------|--------|--------|
+| 0-7 | Active | Normal overview |
+| 8-14 | Cooling | "This project hasn't been touched in [X] days." |
+| 15-30 | Cold | "Cold Project warning. Consider: resume, defer, or archive?" |
+| 30+ | Frozen | "FROZEN PROJECT ([X] days). Archive to `04_Archives/` or reboot?" |
+
+---
+
+## 4. PROCESS NOTES (Absorbed from /project-notes)
+
+**Target:** `project_state.md` > `# Notes`
+
+1. Read all text under `# Notes`.
+2. Identify: concrete action items, bug reports, feature requests.
+3. **Show user proposed additions before writing:** "I found [X] action items in your Notes. Proposed additions to plan.md: [list]. Approve?"
+4. **STOP. Wait for user confirmation.**
+5. After approval: Add to appropriate Phase in `plan.md`.
+6. For architectural changes: Update `## ARCHITECTURE` in `plan.md`.
+7. Mark processed notes: Append `[Processed YYYY-MM-DD]` next to handled notes.
+8. Do NOT delete original text.
+
+---
+
+## 5. OUTPUT THE OVERVIEW
+
+Present a structured summary:
+
+```markdown
+## Project: [Name]
+
+**Status:** [Active/Cooling/Cold/Frozen]
+**Phase:** [Current phase from plan.md]
+**Progress:** [X/Y] tasks ([Z]%)
+**Last Activity:** [Date]
+**Stack:** [Technologies]
+
+### Open Tasks (Top 5)
+1. [ ] [Task] - [Phase]
+2. [ ] [Task] - [Phase]
+
+### Blockers
+- [Blocker 1]
+- [Blocker 2] (or "None")
+
+### Recent Decisions
+- [Decision from changelog]
+
+### Notes Processed
+- [X] notes converted to tasks
+- [Y] notes already processed
+```
+
+---
+
+## 5.5. ACTIVE_CONTEXT SYNC
+
+After presenting the overview, update `Active_Context.md` Section 4 (Technical Context):
+- **Active Project:** [Project name] — [Path]
+- **Tech Stack:** [From brief/plan]
+- **Current Bottleneck:** [Top blocker, or "Execution" if none]
+
+This keeps the state bus current for the next session.
+
+---
+
+## 6. PORTFOLIO MODE (if 'all' requested)
+
+For each project in `01_Projects/`:
+- Read plan.md (if exists). If no plan.md: Status = "Unplanned", Progress = "N/A", Next Action = "Run /ignite or /init-plan"
+- Calculate status and progress
+- Output a summary table:
+
+| Project | Status | Progress | Last Activity | Next Action |
+|---------|--------|----------|---------------|-------------|
+| [Name] | Active | 60% | 2 days ago | [Task] |
+
+Flag any Cold or Frozen projects.
+
+---
+
+## 7. SELF-VERIFICATION (CISC)
+
+Before outputting:
+1. All progress percentages match actual task counts?
+2. Staleness calculation correct?
+3. Notes processed correctly (no data loss)?
+
+If any check fails: flag it.
+
+---
+
+## 8. SELF-IMPROVEMENT
+
+If user corrected AI during this workflow:
+1. Acknowledge: "I made a mistake."
+2. Identify the rule violated.
+3. Append to Auto_Learnings.md:
+   `* [YYYY-MM-DD] CORRECTION: [Description]. Source: User correction. Confidence: 3.`
+4. If same correction seen 3x+, escalate confidence to 5.
+
+---
+
+## 9. PROACTIVE SUGGESTIONS
+
+Based on overview:
+1. If project Cold: "Consider archiving [Project] or scheduling a reboot session." - Confidence: [X]
+2. If high blocker count: "This project has [X] blockers. Prioritize unblocking." - Confidence: [X]
+3. If notes unprocessed: "[X] notes in project_state.md haven't been processed." - Confidence: [X]
+
+"Want me to address any of these?"
+
+---
+
+## 10. ANSWER GATING
+
+Do not suggest handoff until:
+- Project overview has been presented to user
+- Staleness status has been acknowledged
+- Any notes processed are confirmed added to plan.md
+
+If pending: "You haven't confirmed [X]. Wait or proceed?"
+
+---
+
+## HANDOFF
+
+Sources used: [List files read]
+Confidence range: [X-Y across all outputs]
+
+**Next:**
+- If Active project: "Continue working, or run `/log` when you finish a session."
+- If Cold project: "Run `/reboot` to break through, or archive to `04_Archives/`."
+- If notes processed: "New tasks added to plan.md. Start executing."
+
+Key rules still active:
+- All claims must be grounded
+- Confidence < 3 = ask, don't assert
+- Append-only: never delete history