ma-agents 3.4.9 → 3.5.0

package/_bmad-output/implementation-artifacts/done/17-6-done-item-cleanup.md

@@ -0,0 +1,273 @@
+# Story 17.6: Done Item Cleanup
+
+Status: review
+
+<!-- Note: Validation is optional. Run validate-create-story for quality check before dev-story. -->
+
+## Story
+
+As a **Scrum Master**,
+I want done stories and bugs automatically removed from the sprint status and their files moved to a done/ subfolder,
+So that the active sprint view stays clean and the archive is organized.
+
+## Acceptance Criteria
+
+1. **Given** a story or bug status changes to `done`,
+   **When** the sprint status is refreshed (via sprint-planning workflow or sprint-status workflow),
+   **Then** the done item is removed from `sprint-status.yaml`'s `development_status` section,
+   **And** the done item is removed from its sprint file's `items` array,
+   **And** the sprint file's effective capacity increases by 1 (slot freed).
+
+2. **Given** a story file exists at `{implementation_artifacts}/{story-id}.md`,
+   **When** the story reaches `done` status,
+   **Then** the file is moved to `{implementation_artifacts}/done/{story-id}.md`,
+   **And** if the `done/` subfolder doesn't exist, it is created.
+
+3. **Given** a bug file exists at `{implementation_artifacts}/BUG-{title}.md`,
+   **When** the bug reaches `done` status,
+   **Then** the file is moved to `{implementation_artifacts}/done/BUG-{title}.md`.
+   **Note:** Bug files use lowercase `bug-{slug}.md` naming. The file is moved as-is: `done/bug-{slug}.md`.
+
+4. **Given** all items in a sprint reach `done` status,
+   **When** the sprint is refreshed,
+   **Then** the sprint status changes to `closed`,
+   **And** the sprint file is preserved (historical record) but `items` array is empty.
+
+## Tasks / Subtasks
+
+- [x] Task 1: Create cleanup-done SKILL.md (AC: #1, #2, #3, #4)
+  - [x] 1.1: Create skill package at `lib/bmad-extension/skills/cleanup-done/SKILL.md` with YAML frontmatter:
+    ```yaml
+    ---
+    name: cleanup-done
+    description: Archive done stories and bugs — move files to done/ subfolder and remove from sprint/backlog
+    type: skill
+    triggers:
+      - "cleanup done"
+      - "archive done items"
+    ---
+    ```
+  - [x] 1.2: Create `lib/bmad-extension/skills/cleanup-done/bmad-skill-manifest.yaml`
+  - [x] 1.3: Implement workflow steps:
+
+    **Step 1: Identify done items**
+    - Read `_bmad-output/implementation-artifacts/sprint-status.yaml` — find all entries in `development_status` with status `done`
+    - Read `_bmad-output/implementation-artifacts/backlog.yaml` (if exists) — find all entries with `status: done`
+    - Glob `_bmad-output/implementation-artifacts/bug-*.md` — check frontmatter for `status: done`
+    - Build consolidated list of done items from all sources (deduplicate by ID)
+    - If no done items found: display "No done items to clean up." and exit
+
+    **Step 2: Display done items for confirmation**
+    - Show table: #, ID, Title, Type, Current Location, Sprint (if assigned)
+    - Prompt: "[a] Archive all / [s] Select specific items / [q] Cancel"
+    - If 'q': exit
+    - If 's': prompt for item numbers
+
+    **Step 3: Ensure done/ directory exists**
+    - Check if `_bmad-output/implementation-artifacts/done/` exists
+    - If not, create it (AC #2)
+
+    **Step 4: Move files to done/ subfolder (AC: #2, #3)**
+    - For each confirmed done item:
+      - **Stories:** Move `_bmad-output/implementation-artifacts/{story-id}.md` to `_bmad-output/implementation-artifacts/done/{story-id}.md`
+      - **Bugs:** Move `_bmad-output/implementation-artifacts/bug-{slug}.md` to `_bmad-output/implementation-artifacts/done/bug-{slug}.md`
+      - Use file rename (not copy+delete) — preserves git history tracking
+      - If source file doesn't exist (already moved or never created): log warning, skip
+      - If destination file already exists: log warning "File already in done/ archive", skip move
+
+    **Step 5: Remove from sprint files (AC: #1, #4)**
+    - Glob `_bmad-output/implementation-artifacts/sprints/sprint-*.yaml`
+    - For each sprint file:
+      - Read `items` array
+      - Remove any done item IDs from the array
+      - If items were removed: update `last_modified` timestamp, write file
+      - If `items` array is now empty AND sprint status is `active`: change sprint status to `closed` (AC #4)
+      - If `items` array is now empty AND sprint status is `planning`: leave as `planning` (empty planning sprint is valid)
+
+    **Step 6: Remove from backlog.yaml (AC: #1)**
+    - Read `backlog.yaml`
+    - Remove all done items from the backlog list
+    - Re-number priorities as sequential integers 1..N (no gaps)
+    - Write updated backlog.yaml
+
+    **Step 7: Remove from sprint-status.yaml (AC: #1)**
+    - Read `sprint-status.yaml`
+    - For each done story item: remove the key from `development_status` section
+    - Do NOT remove epic keys or retrospective keys — only story/bug keys
+    - If all stories in an epic are now removed (all done): update epic status to `done`
+    - Preserve ALL comments and structure including STATUS DEFINITIONS header
+    - Update `last_updated` field (if it exists)
+    - Write updated file
+
+    **Step 8: Report**
+    - Display summary:
+      - Files moved to done/: count and list
+      - Items removed from sprint files: count
+      - Sprint(s) auto-closed: list (if any)
+      - Items removed from backlog: count
+      - Items removed from sprint-status.yaml: count
+    - Display: "Sprint capacity freed: {count} slot(s)"
+
+- [x] Task 2: Register in module-help.csv (AC: #1)
+  - [x] 2.1: Add row to `lib/bmad-extension/module-help.csv`:
+    ```
+    ma-skills,4-implementation,Cleanup Done,cleanup-done,,skill:cleanup-done,cleanup-done,false,bmm-sm,,"Archive done items — move files to done/ and remove from sprint/backlog.",_bmad-output/implementation-artifacts,"archived items",
+    ```
+
+- [x] Task 3: Add cleanup trigger to sprint-status and sprint-planning workflows
+  - [x] 3.1: Add a note in the cleanup-done SKILL.md output section that sprint-status and sprint-planning workflows should invoke `/cleanup-done` as a pre-step to ensure the view is clean before display/planning
+  - [x] 3.2: Do NOT modify `_bmad/bmm/4-implementation/bmad-sprint-planning/workflow.md` or `_bmad/bmm/4-implementation/bmad-sprint-status/workflow.md` directly — those are bmm core workflows. Instead, document the integration point.
+
+- [x] Task 4: Update guardrails (AC: #1, #2, #3, #4)
+  - [x] 4.1: Add to `sprint-management-guardrails.md`:
+    - Done Item Archive section: files move to `done/` subfolder via rename
+    - Sprint Auto-Close rule: sprint closes when all items are done and removed
+    - Document that cleanup-done is the ONLY workflow that removes items from sprint-status.yaml's development_status (other than manual edits)
+
+- [x] Task 5: Verify (AC: #1, #2, #3, #4)
+  - [x] 5.1: Verify file move uses rename (not copy+delete) for stories and bugs
+  - [x] 5.2: Verify done/ directory is created if missing
+  - [x] 5.3: Verify sprint file items array is updated and last_modified set
+  - [x] 5.4: Verify sprint auto-close when items array becomes empty on active sprint
+  - [x] 5.5: Verify backlog.yaml is cleaned and priorities re-numbered
+  - [x] 5.6: Verify sprint-status.yaml comments and structure are preserved
+  - [x] 5.7: Verify epic status is updated to done when all stories removed
+  - [x] 5.8: Verify existing done/ files are not overwritten (skip with warning)
+
+## Dev Notes
+
+### Architecture Compliance
+
+- **SKILL.md is the authoritative file:** New skill at `lib/bmad-extension/skills/cleanup-done/SKILL.md`
+- **Extension module pattern:** Skill lives in `lib/bmad-extension/skills/cleanup-done/`
+- **Multi-write pattern:** This workflow writes to up to 4 file types: done/ files, sprint files, backlog.yaml, sprint-status.yaml
+- **File rename, not copy+delete:** Use `fs.rename()` equivalent to preserve git history tracking (FR131)
+- **The `done/` subfolder already exists** at `_bmad-output/implementation-artifacts/done/` — this story formalizes its use
+
+### Data Flow
+
+```
+READ:  sprint-status.yaml → identify done story items
+READ:  backlog.yaml → identify done items with backlog entries
+READ:  bug-*.md frontmatter → identify done bugs
+READ:  sprints/sprint-*.yaml → find sprints containing done items
+
+WRITE: done/{file}.md → move file from active to archive
+WRITE: sprints/sprint-{id}.yaml → remove done items from items array
+WRITE: backlog.yaml → remove done items, renumber priorities
+WRITE: sprint-status.yaml → remove done story entries, update epic statuses
+```
+
+### Done Detection Sources
+
+Done items can be detected from multiple sources:
+1. `sprint-status.yaml` development_status entries with value `done` — primary source for stories
+2. `backlog.yaml` entries with `status: done` — if backlog model is active
+3. `bug-*.md` frontmatter with `status: done` — for bugs
+
+The cleanup workflow checks ALL sources and deduplicates by ID to ensure comprehensive cleanup.
+
+### Sprint Auto-Close Logic
+
+```
+IF sprint.items becomes empty after cleanup
+  AND sprint.status == "active"
+THEN sprint.status = "closed"
+
+IF sprint.items becomes empty after cleanup
+  AND sprint.status == "planning"
+THEN no status change (empty planning sprint is valid)
+
+IF sprint.status == "closed"
+THEN skip (already closed, no action needed)
+```
+
+### Epic Auto-Done Logic
+
+```
+IF all story keys matching epic-{N}-* pattern
+  are removed from sprint-status.yaml (all done)
+  AND epic-{N} entry exists
+THEN set epic-{N} status to "done"
+```
+
+### Impact on Other Epic 17 Stories
+
+- **17.1 (Sprint Entity):** Sprint files modified by cleanup (items array, status, last_modified)
+- **17.2 (Flat Backlog):** Backlog.yaml modified by cleanup (done items removed)
+- **17.4 (Backlog-to-Sprint):** After cleanup, freed capacity in sprints becomes available
+- **17.5 (Sprint-to-Backlog):** Different operation — 17.5 returns items to backlog, 17.6 archives them
+- **17.8 (Sprint Status Display):** After cleanup, done items no longer appear in sprint views
+
+### Critical Guardrails (Story-Specific)
+
+- Do NOT modify any files under `_bmad/bmm/workflows/` or `_bmad/core/`
+- Do NOT add npm dependencies
+- Do NOT modify `lib/installer.js`, `lib/agents.js`, or `bin/cli.js`
+- Do NOT delete files — MOVE them to done/ subfolder (FR131)
+- Do NOT overwrite existing files in done/ — skip with warning
+- Do NOT remove epic keys or retrospective keys from sprint-status.yaml — only remove story/bug keys
+- ALWAYS preserve sprint-status.yaml comments and structure
+- ALWAYS use file rename, not copy+delete
+- ALWAYS re-number backlog priorities after removal (no gaps)
+- Sprint auto-close ONLY applies to `active` sprints, not `planning` sprints
+- **Do NOT modify bmm core workflows** — document integration points only
+
+### Project Structure Notes
+
+- **Files to create:**
+  - `lib/bmad-extension/skills/cleanup-done/SKILL.md` — new skill (authoritative)
+  - `lib/bmad-extension/skills/cleanup-done/bmad-skill-manifest.yaml` — skill manifest
+- **Files to modify:**
+  - `lib/bmad-extension/module-help.csv` — add cleanup-done row
+  - `_bmad-output/implementation-artifacts/sprint-management-guardrails.md` — document archive rules, auto-close, auto-done
+- **Files modified at runtime (by the skill):**
+  - `_bmad-output/implementation-artifacts/sprint-status.yaml` — removes done entries
+  - `_bmad-output/implementation-artifacts/backlog.yaml` — removes done items
+  - `_bmad-output/implementation-artifacts/sprints/sprint-*.yaml` — removes done items from arrays
+  - `_bmad-output/implementation-artifacts/*.md` → `done/*.md` — file moves
+- **Files NOT to touch (at development time):**
+  - `_bmad/bmm/4-implementation/bmad-sprint-planning/workflow.md`
+  - `_bmad/bmm/4-implementation/bmad-sprint-status/workflow.md`
+  - Any files under `_bmad/core/` or `lib/installer.js`
+
+### References
+
+- [Source: _bmad-output/planning-artifacts/epics.md#Epic 17, Story 17.6] — Story requirements and acceptance criteria
+- [Source: _bmad-output/planning-artifacts/prd.md#FR130] — Done items removed from sprint status file
+- [Source: _bmad-output/planning-artifacts/prd.md#FR131] — Done story files moved to done/ subfolder
+- [Source: _bmad-output/implementation-artifacts/17-1-sprint-entity-model.md] — Sprint YAML schema, status transitions (active -> closed)
+- [Source: _bmad-output/implementation-artifacts/17-2-flat-backlog-model.md] — Backlog schema (done items excluded)
+- [Source: lib/bmad-extension/skills/create-bug-story/SKILL.md] — Bug file format and naming convention
+- [Source: lib/bmad-extension/module-help.csv] — Workflow registration
+
+### Git Intelligence
+
+Recent commits show Epic 15/16 patterns:
+- Conventional commit format: `feat(epic-17): story 17-6 done item cleanup`
+- New skill creation follows: SKILL.md + bmad-skill-manifest.yaml + module-help.csv registration
+
+## Change Log
+
+- 2026-03-30: Story implemented — created cleanup-done skill with multi-source done detection, file archival, and sprint auto-close.
+
+## Dev Agent Record
+
+### Agent Model Used
+
+Claude Opus 4.6 (1M context)
+
+### Debug Log References
+
+None — clean implementation.
+
+### Completion Notes List
+
+Created cleanup-done skill (SKILL.md + manifest). Registered in module-help.csv. Updated guardrails with archive rules, sprint auto-close, epic auto-done.
+
+### File List
+
+- `lib/bmad-extension/skills/cleanup-done/SKILL.md` (new)
+- `lib/bmad-extension/skills/cleanup-done/bmad-skill-manifest.yaml` (new)
+- `lib/bmad-extension/module-help.csv` (modified)
+- `sprint-management-guardrails.md` (modified)

package/_bmad-output/implementation-artifacts/done/17-7-multi-criteria-prioritization.md

@@ -0,0 +1,235 @@
+# Story 17.7: Multi-Criteria Prioritization Workflow
+
+Status: review
+
+<!-- Note: Validation is optional. Run validate-create-story for quality check before dev-story. -->
+
+## Story
+
+As a **Scrum Master**,
+I want a workflow to reprioritize the backlog using multiple criteria,
+So that sprint planning is based on business value, severity, and dependencies — not just epic order.
+
+## Acceptance Criteria
+
+1. **Given** the backlog contains stories and bugs,
+   **When** the Scrum Master invokes the prioritize-backlog workflow,
+   **Then** it presents each item with its current priority and metadata (type, epic, severity for bugs, dependencies),
+   **And** the Scrum Master can reorder items using criteria: severity, business value, dependencies, item type, age.
+
+2. **Given** the Scrum Master reorders items,
+   **When** the new priority order is confirmed,
+   **Then** `backlog.yaml` is updated with new `priority` values (sequential integers starting from 1),
+   **And** the reordering is logged with date and reason.
+
+3. **Given** items have dependencies (e.g., Story 15.2 depends on 15.1),
+   **When** the Scrum Master tries to prioritize the dependent item above its dependency,
+   **Then** the workflow warns about the dependency conflict,
+   **And** allows the Scrum Master to override or adjust.
+
+## Tasks / Subtasks
+
+- [x] Task 1: Create prioritize-backlog SKILL.md (AC: #1, #2, #3)
+  - [x] 1.1: Create skill package at `lib/bmad-extension/skills/prioritize-backlog/SKILL.md` with YAML frontmatter:
+    ```yaml
+    ---
+    name: prioritize-backlog
+    description: Reprioritize backlog items using multiple criteria — severity, business value, dependencies, type, age
+    type: skill
+    triggers:
+      - "prioritize backlog"
+      - "reorder backlog"
+      - "reprioritize"
+    ---
+    ```
+  - [x] 1.2: Create `lib/bmad-extension/skills/prioritize-backlog/bmad-skill-manifest.yaml`
+  - [x] 1.3: Implement workflow steps:
+
+    **Step 1: Load backlog and infer dependencies**
+    - Read `_bmad-output/implementation-artifacts/backlog.yaml`
+    - If no backlog file: display "No backlog.yaml found. Run `/generate-backlog` first." and exit
+    - If backlog is empty: display "Backlog is empty." and exit
+    - Infer dependencies from epic story ordering:
+      - Stories within the same epic are sequentially dependent by default (e.g., 17-1 before 17-2 before 17-3)
+      - Cross-epic dependencies are inferred from epics.md dependency notes if available
+    - Read `_bmad-output/planning-artifacts/epics.md` for explicit dependency information
+
+    **Step 2: Display current backlog with metadata**
+    - Show table with columns: #, Priority, ID, Title, Type, Epic, Status, Sprint, Severity, Dependencies
+    - For bugs: show severity in its own column
+    - For stories: show epic number
+    - For items with dependencies: show which items they depend on (e.g., "after 17-1")
+    - For sprint-assigned items: show sprint ID
+    - Color/marker coding: unassigned items vs sprint-assigned items
+
+    **Step 3: Choose prioritization mode**
+    - Prompt: "How would you like to prioritize?"
+    - Options:
+      1. **Full reprioritization** — present all items, reorder from scratch
+      2. **Quick adjust** — move specific items up or down
+      3. **Auto-suggest** — AI recommends optimal ordering based on criteria
+      4. **Cancel** — exit without changes
+
+    **Step 3a: Full reprioritization (if mode = 1)**
+    - Prompt: "Enter item numbers in your desired priority order (comma-separated). Items not listed will be appended at the end in their current order."
+    - Validate all numbers are valid item references
+    - Check for dependency violations (AC #3)
+    - Build new ordering: user-specified items first, remaining items appended
+
+    **Step 3b: Quick adjust (if mode = 2)**
+    - Prompt: "Enter item number to move:"
+    - Prompt: "Move to position (number) or relative move ([u]p N / [d]own N):"
+    - Check for dependency violations at new position
+    - Apply move and re-number priorities
+    - Show updated backlog; prompt: "[m] Move another / [d] Done"
+
+    **Step 3c: Auto-suggest (if mode = 3)**
+    - Analyze all items using weighted criteria:
+      - **Blocking dependencies first** (items that unblock others get highest priority)
+      - **Critical/High bugs** above medium/low stories (severity weighting)
+      - **Business value** (inferred from epic priority and story position)
+      - **Sprint-assigned items** grouped together (don't interleave assigned/unassigned unnecessarily)
+      - **Effort** (smaller items above larger when other criteria are equal)
+    - Present suggested reordering with rationale for each major change
+    - Prompt: "[a] Accept suggestion / [m] Modify / [r] Reject and keep current"
+
+    **Step 4: Dependency validation (AC: #3)**
+    - For the proposed new ordering, check each item:
+      - If item X depends on item Y, verify Y has a lower priority number (higher priority) than X
+      - If violation found: display warning:
+        ```
+        Warning: "{item_X}" depends on "{item_Y}", but {item_X} is prioritized above {item_Y}.
+        [o] Override — I know what I'm doing
+        [f] Fix — move {item_X} below {item_Y}
+        [c] Cancel reordering
+        ```
+
+    **Step 5: Confirm and persist (AC: #2)**
+    - Display the new ordering side-by-side with old ordering (show what changed)
+    - Prompt: "Reason for reprioritization (brief description):"
+    - Store reason for audit log
+    - Prompt: "[y] Confirm / [n] Cancel / [e] Edit"
+    - If confirmed:
+      - Update `backlog.yaml` with new `priority` values (sequential 1..N)
+      - Append to backlog.yaml header comment: `# Reprioritized: {date} — {reason}`
+      - Write updated file
+    - Display: "Backlog reprioritized. {count} items reordered."
+
+- [x] Task 2: Register in module-help.csv (AC: #1)
+  - [x] 2.1: Add row to `lib/bmad-extension/module-help.csv`:
+    ```
+    ma-skills,4-implementation,Prioritize Backlog,prioritize-backlog,,skill:prioritize-backlog,prioritize-backlog,false,bmm-sm,,"Reprioritize backlog using multiple criteria — severity, value, dependencies.",_bmad-output/implementation-artifacts,"backlog",
+    ```
+
+- [x] Task 3: Update guardrails (AC: #2, #3)
+  - [x] 3.1: Add to `sprint-management-guardrails.md`:
+    - Prioritization Rules section: criteria list, dependency validation, audit logging
+    - Document that priority values are always sequential 1..N after any reordering
+    - Document dependency inference: same-epic stories are sequentially dependent by default
+
+- [x] Task 4: Verify (AC: #1, #2, #3)
+  - [x] 4.1: Trace SKILL.md — confirm all three modes (full, quick, auto-suggest) produce valid orderings
+  - [x] 4.2: Verify priority re-numbering produces sequential integers 1..N with no gaps
+  - [x] 4.3: Verify dependency violation detection and warning display
+  - [x] 4.4: Verify override capability (Scrum Master can override dependency warnings)
+  - [x] 4.5: Verify audit logging (date + reason appended to backlog.yaml header)
+  - [x] 4.6: Verify sprint-assigned items' `sprint` field is preserved during reordering
+
+## Dev Notes
+
+### Architecture Compliance
+
+- **SKILL.md is the authoritative file:** New skill at `lib/bmad-extension/skills/prioritize-backlog/SKILL.md`
+- **Extension module pattern:** Skill lives in `lib/bmad-extension/skills/prioritize-backlog/`
+- **Single-file write:** Only writes to `backlog.yaml` — no other files modified
+- **Non-destructive:** Reordering only changes `priority` values, never changes `status`, `sprint`, or other fields
+
+### Prioritization Criteria (Weighted)
+
+| Criterion | Weight | Description |
+|---|---|---|
+| Blocking dependencies | Highest | Items that unblock others get top priority |
+| Severity (bugs) | High | Critical > High > Medium > Low |
+| Business value | Medium | Inferred from epic priority and story position |
+| Item type | Low | Bugs slightly above stories when severity/value equal |
+| Age | Lowest | Older items above newer when all else equal |
+
+### Dependency Inference Rules
+
+1. **Same-epic sequential:** Story N depends on Story N-1 within the same epic (e.g., 17-3 depends on 17-2)
+2. **Cross-epic explicit:** Dependencies stated in epics.md "Dependencies" sections (e.g., Epic 17 depends on Epic 11)
+3. **Override allowed:** Scrum Master can override any dependency warning — the workflow warns but doesn't block
+
+### Impact on Other Epic 17 Stories
+
+- **17.2 (Flat Backlog):** Provides the backlog.yaml this story reads/writes
+- **17.4 (Backlog-to-Sprint):** Reads backlog.yaml ordered by priority set by this story
+- **17.8 (Sprint Status Display):** Displays backlog items in priority order set by this story
+
+### Critical Guardrails (Story-Specific)
+
+- Do NOT modify any files under `_bmad/bmm/workflows/` or `_bmad/core/`
+- Do NOT add npm dependencies
+- Do NOT modify `lib/installer.js`, `lib/agents.js`, or `bin/cli.js`
+- Do NOT modify `sprint-status.yaml` — priorities are in backlog.yaml only
+- Do NOT change any field other than `priority` when reordering
+- ALWAYS re-number priorities as sequential integers 1..N (no gaps, no ties)
+- ALWAYS preserve `sprint`, `status`, `severity`, and all other fields during reordering
+- ALWAYS validate dependencies before confirming (but allow override)
+- ALWAYS log the reordering reason and date in backlog.yaml header comment
+
+### Project Structure Notes
+
+- **Files to create:**
+  - `lib/bmad-extension/skills/prioritize-backlog/SKILL.md` — new skill (authoritative)
+  - `lib/bmad-extension/skills/prioritize-backlog/bmad-skill-manifest.yaml` — skill manifest
+- **Files to modify:**
+  - `lib/bmad-extension/module-help.csv` — add prioritize-backlog row
+  - `_bmad-output/implementation-artifacts/sprint-management-guardrails.md` — document prioritization rules
+- **Files modified at runtime:**
+  - `_bmad-output/implementation-artifacts/backlog.yaml` — priority values updated
+- **Files NOT to touch:**
+  - `sprint-status.yaml` (priorities not stored here)
+  - Sprint files (priorities not stored here)
+  - Any files under `_bmad/` or `lib/installer.js`
+
+### References
+
+- [Source: _bmad-output/planning-artifacts/epics.md#Epic 17, Story 17.7] — Story requirements and acceptance criteria
+- [Source: _bmad-output/planning-artifacts/prd.md#FR67] — Multi-criteria prioritization beyond epic precedence
+- [Source: _bmad-output/planning-artifacts/prd.md#FR65] — Flat backlog with priority independent of epic
+- [Source: _bmad-output/implementation-artifacts/17-2-flat-backlog-model.md] — Backlog YAML schema with priority field
+- [Source: _bmad-output/implementation-artifacts/17-4-backlog-to-sprint-workflow.md] — Consumes prioritized backlog
+- [Source: lib/bmad-extension/skills/add-to-sprint/SKILL.md] — Current multi-criteria analysis (moved to dedicated skill)
+- [Source: lib/bmad-extension/module-help.csv] — Workflow registration
+
+### Git Intelligence
+
+Recent commits show Epic 15/16 patterns:
+- Conventional commit format: `feat(epic-17): story 17-7 multi-criteria prioritization`
+- New skill creation follows: SKILL.md + bmad-skill-manifest.yaml + module-help.csv registration
+
+## Change Log
+
+- 2026-03-30: Story implemented — created prioritize-backlog skill with full/quick/auto-suggest modes, dependency inference, and audit logging.
+
+## Dev Agent Record
+
+### Agent Model Used
+
+Claude Opus 4.6 (1M context)
+
+### Debug Log References
+
+None — clean implementation.
+
+### Completion Notes List
+
+Created prioritize-backlog skill (SKILL.md + manifest). Registered in module-help.csv. Updated guardrails with prioritization rules, dependency inference, audit logging.
+
+### File List
+
+- `lib/bmad-extension/skills/prioritize-backlog/SKILL.md` (new)
+- `lib/bmad-extension/skills/prioritize-backlog/bmad-skill-manifest.yaml` (new)
+- `lib/bmad-extension/module-help.csv` (modified)
+- `sprint-management-guardrails.md` (modified)