@hanzlaa/rcode 3.6.16 → 4.0.0

package/rcode/workflows/create-epics-and-stories.md

@@ -0,0 +1,388 @@
+# Workflow: rcode-create-epics-and-stories
+
+<purpose>
+Parse a PRD, PROJECT.md, or project document to generate numbered epic files in `.planning/epics/`. Each epic file contains user stories with acceptance criteria, development notes, and effort estimates. Output is ready for `/rcode-sprint-planning`.
+</purpose>
+
+
+<available_agent_types>
+- `rcode-roadmapper` — reads PRD/context and generates epic structure
+</available_agent_types>
+
+## Step 0 — Parse Arguments & Detect Context Mode
+
+```bash
+INPUT_FILE=""
+PREFIX=""
+CONTEXT_MODE="file"   # file | interactive | codebase
+
+for arg in $ARGUMENTS; do
+  if [[ "$arg" == "--prefix" ]]; then PREFIX_NEXT=true
+  elif [[ "$PREFIX_NEXT" == true ]]; then PREFIX="$arg"; PREFIX_NEXT=false
+  elif [[ "$arg" != --* ]]; then INPUT_FILE="$arg"
+  fi
+done
+```
+
+**Route based on what was provided:**
+
+| Situation | Action |
+|-----------|--------|
+| `INPUT_FILE` given and file exists | Use it — `CONTEXT_MODE=file` |
+| `INPUT_FILE` given but file missing | Error — "File not found: {path}" — STOP |
+| No `INPUT_FILE` given, `.planning/PROJECT.md` exists | Use PROJECT.md — `CONTEXT_MODE=file` |
+| No `INPUT_FILE` given, no PROJECT.md | Ask user (see below) |
+
+**When no context file exists — ask via AskUserQuestion:**
+
+```
+No PRD or PROJECT.md found. How should we proceed?
+
+1. Initialize the project first — run /rcode-new-project to capture goals, stack, 
+   milestones, and generate PROJECT.md. Recommended if this is a new project.
+
+2. Gather context from codebase + GitHub issues — I'll scan the repo, read open 
+   issues, and use that as the foundation. Good for existing projects.
+
+3. Tell me what you want to build — describe the feature/epic scope and I'll 
+   generate stories from that description directly.
+
+0. Cancel — exit without changes.
+```
+
+- If **1**: invoke `/rcode-new-project` and stop.
+- If **2**: set `CONTEXT_MODE=codebase`
+- If **3**: capture description → set `CONTEXT_MODE=interactive`, `DESCRIPTION=$response`
+- If **0**: print `Cancelled.` and STOP.
+
+## Step 1 — Load References & Gather Deep Context
+
+```bash
+@.rcode/references/checklist-story-draft.md
+@.rcode/references/commit-conventions.md
+```
+
+**Context gathering varies by mode:**
+
+### Mode: `file`
+
+Read `INPUT_FILE`. Also run supplemental scan:
+
+```bash
+STACK=$(cat .planning/STACK.md 2>/dev/null || cat README.md 2>/dev/null | head -60)
+OPEN_ISSUES=$(gh issue list --state open --limit 50 --json number,title,labels,assignees 2>/dev/null)
+RECENT_COMMITS=$(git log --oneline -20 2>/dev/null)
+```
+
+### Mode: `codebase`
+
+Do NOT proceed with thin analysis. Run deep parallel scan:
+
+```bash
+# 1. Codebase structure
+CODEBASE=$(gemini -p "@./ Summarize: tech stack, main modules, key entry points, 
+  any existing feature flags or TODOs. Under 400 words.")
+
+# 2. Open GitHub issues (grouped by label/theme)
+OPEN_ISSUES=$(gh issue list --state open --limit 100 \
+  --json number,title,labels,assignees,body 2>/dev/null)
+
+# 3. Recent git activity (last 30 commits = what changed recently)
+RECENT_WORK=$(git log --oneline -30 2>/dev/null)
+
+# 4. Existing planning files
+EXISTING_EPICS=$(ls .planning/epics/*.md 2>/dev/null | xargs grep '^# Epic' 2>/dev/null)
+```
+
+After gathering, synthesize into a brief context document:
+
+```
+Project context:
+- Stack: {from codebase scan}
+- Recent work: {from git log}
+- Open issues: {grouped themes from GitHub}
+- Existing epics: {if any}
+- User's scope description: {from ARGUMENTS or interactive input}
+```
+
+### Mode: `interactive`
+
+Use the user's description + quick codebase scan:
+
+```bash
+CODEBASE_QUICK=$(gemini -p "@./ In 200 words: tech stack and main modules only.")
+```
+
+## Step 2 — Generate Epic Structure
+
+Call `rcode-roadmapper` with full context. The prompt MUST enforce proper epic decomposition:
+
+```
+You are a senior product manager breaking down a feature area into epics and stories.
+
+## Context
+
+{Full context document from Step 1 — PRD/PROJECT.md content, codebase summary, 
+open issues, recent git activity}
+
+## Your task
+
+Generate a structured set of epics. Each epic covers ONE coherent concern or phase.
+
+## Epic decomposition rules (MANDATORY)
+
+1. One epic = one coherent theme. Examples of correct splits:
+   - ✓ "Web Search Investigation" + "Toggle Bug Fixes" + "Provider Improvements" = 3 epics
+   - ❌ "Web Search (everything)" with 10 stories = 1 bloated epic (WRONG)
+   
+2. Each epic has 3–5 stories MAX. If you have 8+ stories for one theme, split into 2 epics.
+
+3. Epics are ordered by dependency: investigation before implementation, 
+   infrastructure before features, blockers before dependent work.
+
+4. Story IDs use plain numbers: 1.1, 1.2 — NOT "EPIC-1.1". The epic number is 
+   implied by the file. Story IDs must never repeat across epics.
+
+5. A story must be completable in 1–5 days (S or M effort). L effort stories 
+   must be split into smaller stories.
+
+6. Investigation/spike stories are valid and should be their own epic when 
+   there are 3+ unknowns to resolve.
+
+## Output format (strict JSON)
+
+{
+  "epics": [
+    {
+      "number": 1,
+      "title": "Short epic title (max 6 words)",
+      "description": "2-3 sentences: what this epic covers, why it exists, what done looks like",
+      "phase": "investigation | design | implementation | testing | release",
+      "depends_on": [],
+      "stories": [
+        {
+          "number": 1,
+          "title": "Story title (max 8 words)",
+          "persona": "Full name + role (e.g. 'Hanzla (Backend Engineer)')",
+          "action": "I want to {specific action}",
+          "outcome": "so that {specific measurable outcome}",
+          "acceptance_criteria": [
+            "Specific, testable criterion 1",
+            "Specific, testable criterion 2",
+            "Specific, testable criterion 3"
+          ],
+          "out_of_scope": ["What this story explicitly does NOT cover"],
+          "effort": "S|M|L",
+          "effort_rationale": "why this size — name the files/systems involved",
+          "dev_notes": "Exact file paths, function names, line-level hints. Be specific.",
+          "linked_issues": ["#123", "#456"]
+        }
+      ]
+    }
+  ]
+}
+```
+
+## Quality checks
+
+- Reject epics with >5 stories — split them
+- Reject stories with vague acceptance criteria ("it works", "it is fast")  
+- Reject stories missing dev_notes (every story needs at least 2 file paths)
+- Reject L-effort stories — split into 2 M stories
+- linked_issues must reference real issue numbers from the context above
+```
+
+Wait for the roadmapper response (JSON).
+
+**Validate response before continuing:**
+
+```bash
+EPIC_COUNT=$(echo "$EPIC_JSON" | jq '.epics | length')
+MAX_STORIES=$(echo "$EPIC_JSON" | jq '[.epics[].stories | length] | max')
+
+if [[ $MAX_STORIES -gt 5 ]]; then
+  echo "⚠ Epic has >5 stories — asking roadmapper to split..."
+  # Re-prompt: "Epic {N} has {count} stories. Split into 2 epics by grouping 
+  # closely related stories. Return updated JSON."
+fi
+
+if [[ $EPIC_COUNT -lt 2 && $MAX_STORIES -gt 4 ]]; then
+  echo "⚠ Only 1 epic generated — likely under-split. Asking roadmapper to separate concerns..."
+fi
+```
+
+## Step 3 — Validate Epic Structure
+
+Validate each story passes draft checklist before writing any files:
+- Persona named ✓
+- Action/outcome specified ✓
+- 3+ acceptance criteria ✓
+- Out-of-scope listed ✓
+- Effort is S or M (never L) ✓
+- dev_notes contains at least 2 file paths ✓
+
+If any story fails, ask roadmapper to fix before continuing.
+
+## Step 4 — Write Split Files
+
+**Layout:**
+```
+.planning/epics/
+  EPIC-01.md          ← lean summary only (~25 lines)
+  EPIC-02.md
+  INDEX.md            ← master table of all epics + stories
+  stories/
+    1.1.md            ← one file per story (~35 lines)
+    1.2.md
+    2.1.md
+```
+
+**Why split:** Epic files are kept lean so GitHub API updates only rewrite one small story file instead of the full epic. Token cost per `/rcode-dev-story` is proportional to one story, not the whole epic.
+
+```bash
+mkdir -p .planning/epics/stories
+```
+
+**For each epic — write lean `EPIC-{NN}.md`:**
+
+```markdown
+# Epic {N}: {Title}
+
+**Phase:** {phase}
+**Scope:** {2-3 sentence description}
+**Depends on:** {epic numbers or "none"}
+
+## Stories
+
+| Story | Title | Effort | Status |
+|-------|-------|--------|--------|
+| {N}.1 | {title} | S | — |
+| {N}.2 | {title} | M | — |
+
+**Total:** {sum effort}
+
+---
+*Stories: `.planning/epics/stories/{N}.1.md` … `{N}.{last}.md`*
+```
+
+**For each story — write `.planning/epics/stories/{N}.{M}.md`:**
+
+```markdown
+# Story {N}.{M}: {Title}
+
+**Epic:** EPIC-{N} — {Epic title}
+**Persona:** {Full name + role}
+**Effort:** {S | M}
+**Status:** todo
+**Linked issues:** {#123, #456 or none}
+
+---
+
+## User Story
+
+As a {persona}, I want to {action} so that {outcome}.
+
+## Acceptance Criteria
+
+- [ ] {Criterion 1}
+- [ ] {Criterion 2}
+- [ ] {Criterion 3}
+
+## Out of Scope
+
+- {Exclusion} — {why}
+
+## Dev Notes
+
+{Exact file paths, function names, line-level hints}
+
+---
+*Start: `/rcode-dev-story {N}.{M}`*
+```
+
+## Step 5 — Write INDEX.md
+
+`.planning/epics/INDEX.md` — master overview, kept under 60 lines:
+
+```markdown
+# Epic Index
+
+Generated: {ISO date} | Source: {input filename}
+
+## Epics
+
+| Epic | Title | Phase | Stories | Effort | Depends on |
+|------|-------|-------|---------|--------|------------|
+| EPIC-01 | {title} | investigation | 4 | ~1.5w | — |
+| EPIC-02 | {title} | implementation | 3 | ~2w | EPIC-01 |
+
+**Total effort:** {sum}
+
+## All Stories
+
+| Story | Title | Effort | Status | Issues |
+|-------|-------|--------|--------|--------|
+| 1.1 | {title} | S | todo | #1959 |
+| 1.2 | {title} | M | todo | #1909 |
+| 2.1 | {title} | M | todo | — |
+
+---
+*Update story status: edit `.planning/epics/stories/{N}.{M}.md` → change `Status:` line*
+*Next: `/rcode-sprint-planning` or `/rcode-dev-story {N}.{M}`*
+```
+
+## Step 6 — Commit
+
+```bash
+git add .planning/epics/
+git commit -m "feat(epics): generate epic structure from $(basename $INPUT_FILE)"
+```
+
+Print summary:
+
+```
+Epic structure generated
+
+Epics:   {count} files in .planning/epics/
+Stories: {count} files in .planning/epics/stories/
+Index:   .planning/epics/INDEX.md
+Effort:  ~{weeks} total
+
+Next:
+  /rcode-sprint-planning          organise into sprints
+  /rcode-dev-story {N}.{M}        start working on a story
+```
+
+## Errors
+
+- **Input file not found:** print error and exit
+- **Roadmapper returns invalid JSON:** ask roadmapper to fix and re-run
+- **Fewer than 3 epics:** warn and ask roadmapper to expand
+- **Story missing acceptance criteria:** mark with ⚠️ and ask roadmapper to add
+
+## Success Criteria
+
+- [ ] At least one epic file written to `.planning/epics/EPIC-NN.md` with a non-empty stories list
+- [ ] Each epic file contains user stories with acceptance criteria, dev notes, and effort estimates
+- [ ] Epic files numbered sequentially (`EPIC-01`, `EPIC-02`, …) with no gaps
+- [ ] Story IDs follow `EPIC-NN.M` format (e.g., `EPIC-01.1`, `EPIC-01.2`)
+- [ ] No placeholder text remains — output is ready for `/rcode-sprint-planning` consumption
+
+## On Error
+
+If arguments are invalid, missing files, or subagent fails:
+- Validate inputs match expected format
+- Check that required files exist
+- Retry with clearer arguments or report the specific error to the user
+
+
+## ▶ Next Up
+
+- /rcode-sprint-planning
+- /rcode-dev-story {story-id}
+- /rcode-edit-prd
+
+## Next Up
+
+- `/rcode-sprint-planning` — plan the sprint from the generated epic files
+- `/rcode-create-story` — develop individual stories into self-contained STORY.md files

package/rcode/workflows/create-prd.md

@@ -0,0 +1,30 @@
+# Workflow: rcode-create-prd
+
+<purpose>
+Create a Product Requirements Document from scratch through guided facilitation. Delegates to the rcode-create-prd skill.
+</purpose>
+
+## Execution
+
+Locate and follow the installed skill:
+
+```bash
+find .rcode/skills/actions -path "*rcode-create-prd/workflow.md" 2>/dev/null | head -1
+```
+
+Read and follow the workflow at that path. If the path is empty:
+
+```
+Skill not installed — run: npx @hanzlaa/rcode install
+```
+
+## On Completion
+
+/rcode-validate-prd — validate the PRD for completeness
+/rcode-new-milestone — build the milestone roadmap from the PRD
+/rcode-edit-prd — revise the PRD
+
+## Next Up
+
+- `/rcode-validate-prd` — validate the new PRD for completeness and testability
+- `/rcode-create-architecture` — design the system architecture once the PRD is approved

package/rcode/workflows/create-story.md

@@ -0,0 +1,303 @@
+# Workflow: rcode-create-story
+
+<purpose>
+Convert a single story from an EPIC file into a self-contained STORY.md file. This story is ready for `/rcode-dev-story` to be wrapped for AI-coder execution. Entry is gated by checklist-story-draft.md.
+</purpose>
+
+
+## Step 0 — Usage check
+
+If `$ARGUMENTS` is empty or contains only `--help` or `-h`:
+
+```
+/rcode-create-story <argument-here>
+```
+
+**Examples:**
+```
+/rcode-create-story example 1
+/rcode-create-story example 2
+```
+
+STOP — do not proceed.
+
+## Step 0 — Validation
+
+**If no arguments:**
+
+```
+Usage: /rcode-create-story <EPIC-file.md> [--story <id>]
+
+Examples:
+  /rcode-create-story .planning/epics/EPIC-01.md
+  /rcode-create-story .planning/epics/EPIC-01.md --story EPIC-01.1
+```
+
+Stop and wait for arguments.
+
+**Validate input file exists:**
+
+```bash
+if [[ ! -f "$EPIC_FILE" ]]; then
+  echo "Error: File not found: $EPIC_FILE"
+  exit 1
+fi
+```
+
+**Parse flags:**
+
+- `--story` (optional): specific story ID to extract (e.g., `EPIC-01.1`). If omitted, list stories and ask user to pick one.
+
+## Step 1 — List Stories & Pick One
+
+Read the EPIC file and extract all story IDs and titles:
+
+```bash
+STORIES=$(grep -E "^## Story" "$EPIC_FILE" | sed 's/^## Story //' | head -1 -c 20)
+```
+
+If `--story` flag not provided, ask the user to pick:
+
+```
+Which story from this epic?
+
+- EPIC-01.1: {Title}
+- EPIC-01.2: {Title}
+- EPIC-01.3: {Title}
+
+Select by ID:
+```
+
+Parse the selected story ID.
+
+## Step 2 — Extract Story Content
+
+Read the EPIC file and extract the chosen story:
+
+```bash
+STORY_ID="EPIC-01.1"
+STORY_SECTION=$(sed -n "/^## Story $STORY_ID/,/^## Story /p" "$EPIC_FILE" | head -n -1)
+```
+
+Extract fields from the story markdown:
+
+```bash
+STORY_TITLE=$(echo "$STORY_SECTION" | grep "^## Story" | sed "s/^## Story [^ ]* //")
+PERSONA=$(echo "$STORY_SECTION" | grep "^**Persona:**" | sed "s/^**Persona:** //")
+ACTION=$(echo "$STORY_SECTION" | grep "^**Action:**" | sed "s/^**Action:** //")
+CRITERIA=$(echo "$STORY_SECTION" | sed -n '/^### Acceptance Criteria/,/^### /p' | grep "^- " | sed 's/^- \[ \] //')
+OUT_OF_SCOPE=$(echo "$STORY_SECTION" | sed -n '/^### Out of Scope/,/^### /p' | grep "^- " | sed 's/^- //')
+EFFORT=$(echo "$STORY_SECTION" | grep "^**Estimate:**" | sed "s/^**Estimate:** //")
+RATIONALE=$(echo "$STORY_SECTION" | grep -A1 "^**Rationale:**" | tail -1)
+DEV_NOTES=$(echo "$STORY_SECTION" | sed -n '/^### Dev Notes/,/^---$/p' | sed '1d;$d')
+```
+
+## Step 3 — Load References
+
+```bash
+@.rcode/references/checklist-story-draft.md
+@.rcode/references/checklist-story-dod.md
+@.rcode/references/commit-conventions.md
+```
+
+## Step 4 — Check Entry Gate
+
+Run story through checklist-story-draft.md validation:
+
+- ✓ Persona named
+- ✓ Action/outcome specified
+- ✓ 3+ acceptance criteria
+- ✓ Out-of-scope listed
+- ✓ Effort estimated
+
+If any check fails, print error and stop:
+
+```
+⚠️ Story does not pass draft checklist:
+
+Missing: {failed checks}
+
+Return to the EPIC file and fix before creating STORY.md.
+```
+
+Only proceed if ALL checks pass.
+
+## Step 5 — Create STORY.md
+
+Create `.planning/stories/{story-id}.md`:
+
+```bash
+mkdir -p .planning/stories
+STORY_FILE=".planning/stories/${STORY_ID}.md"
+```
+
+Generate file with template:
+
+```markdown
+# Story: {Title}
+
+**Epic:** {EPIC-XX} — {Epic title}
+
+**ID:** {STORY_ID}
+
+**Status:** Draft
+
+---
+
+## User Story
+
+**Persona:** {Named persona with role}
+
+**Action:** {As a X, I want to Y so that Z}
+
+**Value:** {Expected business impact}
+
+---
+
+## Acceptance Criteria
+
+- [ ] {Criterion 1 — testable condition}
+- [ ] {Criterion 2 — testable condition}
+- [ ] {Criterion 3 — testable condition}
+
+Each criterion can be verified without implementation ambiguity.
+
+---
+
+## Out of Scope
+
+- {What this story does NOT do}
+- {What's in a future story}
+
+Prevents scope creep during development.
+
+---
+
+## Effort Estimate
+
+**Size:** {S | M | L}
+
+**Why:** {Rationale from epic}
+
+**Confidence:** {High | Medium | Low}
+
+---
+
+## Dev Notes
+
+### Technical Approach
+
+{Implementation hints from epic}
+
+### Known Risks
+
+- {Risk 1 and mitigation}
+- {Risk 2 and mitigation}
+
+### Dependencies
+
+- {What must be done first}
+- {External blockers}
+
+### Testing Strategy
+
+From **Definition of Done**:
+- [ ] All acceptance criteria tested and passing
+- [ ] Edge cases: {list specific edge cases relevant to this story}
+- [ ] Tests written: {unit tests for X, integration tests for Y}
+- [ ] Coverage: {target coverage %}
+
+---
+
+## Definition of Done
+
+Use checklist-story-dod.md. A story is "Done" when:
+
+### Acceptance Criteria Verification
+- [ ] All {count} acceptance criteria met
+- [ ] Edge cases handled
+- [ ] No known stubs or TODOs left
+
+### Testing
+- [ ] Tests written for all new logic
+- [ ] All tests passing
+- [ ] Coverage adequate (80%+ on modified code)
+
+### Code Quality
+- [ ] No lint errors
+- [ ] Follows project patterns
+- [ ] Imports all correct
+
+### Commits
+- [ ] Conventional Commits format
+- [ ] No AI attribution
+- [ ] Meaningful messages
+
+### Handoff
+- [ ] Story summary written
+- [ ] No blockers remain
+
+---
+
+## Workflow
+
+**Before you start:** Review this story, dev notes, and dependencies.
+
+**During development:** See `/rcode-dev-story` to wrap this STORY.md for AI-coder execution.
+
+**After complete:** Run checklist-story-dod.md to verify done.
+
+---
+
+**Created:** {ISO date}
+
+**Ready for:** /rcode-dev-story → /rcode → Definition of Done verification
+```
+
+## Step 6 — Commit Story File
+
+```bash
+git add "$STORY_FILE"
+git commit -m "docs(story): create story $STORY_ID from epic"
+```
+
+Print:
+
+```
+📖 Story file created
+
+Location: {path to STORY.md}
+ID: {STORY_ID}
+Title: {Story title}
+Effort: {Estimate}
+Persona: {Persona}
+
+Next step: /rcode-dev-story .planning/stories/{story-id}.md
+```
+
+## Errors
+
+- **Epic file not found:** print error and exit
+- **Story ID not found in epic:** print available IDs and ask user to pick again
+- **Story fails draft checklist:** print failures and ask user to return to epic
+
+## Success Criteria
+
+- [ ] `STORY.md` written to `.planning/stories/{story-id}.md` with title, acceptance criteria, dev notes, effort estimate, and parent epic reference
+- [ ] Entry gate (`checklist-story-draft.md`) passed before the story file is created
+- [ ] Story ID matches the source epic entry in `EPIC-NN.M` format
+- [ ] Story committed to git with a conventional commit message referencing the story ID
+- [ ] Story file is ready for `/rcode-dev-story` consumption without further editing
+
+## On Error
+
+If arguments are invalid, missing files, or subagent fails:
+- Validate inputs match expected format
+- Check that required files exist
+- Retry with clearer arguments or report the specific error to the user
+
+
+## Next Up
+
+- `/rcode-dev-story` — wrap the story for AI-coder execution
+- `/rcode-plan` — plan the phase that will implement this story