valent-pipeline 0.2.21 → 0.2.23

package/pipeline/steps/orchestration/sprint-review.md

@@ -14,13 +14,14 @@ Read each story's `phase-timing.md` for `total_elapsed_minutes` and count reject
 ## Step 2: Fill Sprint Summary
 
 Update `sprint-{n}-plan.md` Sprint Summary:
-- Stories shipped: count and list
+- Stories shipped: count and list (distinguish originally planned vs mid-sprint pulls)
 - Stories rolled over: count and list (if any)
-- Points shipped: sum of shipped story points
+- Points shipped: sum of shipped story points (includes mid-sprint pulls)
 - Points rolled over: sum of unexecuted story points
 - Total elapsed minutes: sum of execution time (not grooming)
-- Velocity this sprint: points_shipped
+- Velocity this sprint: points_shipped (all shipped stories count toward velocity, including pulls)
 - Updated velocity (SMA-5): compute new moving average
+- Mid-sprint pulls: count and list (stories pulled from groomed buffer during execution)
 
 ## Step 3: Finalize Sprint Status YAML
 
@@ -60,6 +61,14 @@ Spawn the Retrospective Agent with this sprint's shipped stories. Pass:
 
 After retrospective completes, update `pipeline-state.json`: `current_sprint.phase = "completed"`.
 
+## Step 6b: Spawn PM (if enabled)
+
+If `{pm_enabled}` is true and the sprint loop will continue (unshipped stories remain in the backlog):
+
+Read and follow `.valent-pipeline/steps/orchestration/sprint-pm-review.md`.
+
+PM spawns concurrently with any remaining retrospective wrap-up. TeamLead waits for PM's `[BACKLOG-UPDATE]` before proceeding to next sprint-init. PM stays alive into the next sprint's grooming phase for FR-tag oversight.
+
 ## Step 7: Check for Next Sprint
 
 If unshipped stories remain in `{backlog_path}` (status `pending` with dependencies met):

package/skills/valent-pm/SKILL.md

@@ -0,0 +1,121 @@
+---
+name: valent-pm
+description: 'Queue a message for the PM agent — product ideas, priority changes, or questions about PRD coverage. Use when the user says "/pm" followed by their message.'
+argument-hint: '<message>'
+---
+
+# valent-pm
+
+Queue a message for the PM (Product Manager) agent. PM processes messages when it next spawns (between sprints). Priority overrides take effect immediately.
+
+## Arguments
+
+The user provides a free-form message after `/pm`. Examples:
+
+- `/pm I want to add keyboard shortcuts for all board actions`
+- `/pm CSV export should also support JSON format`
+- `/pm Reprioritize KANBAN-015 above KANBAN-010, the sync feature is more critical`
+- `/pm What's the current PRD coverage?`
+
+If no argument is provided, ask the user what they'd like to tell PM.
+
+## Execution Steps
+
+### Step 1: Load Pipeline Config
+
+Read `.valent-pipeline/pipeline-config.yaml`. Resolve:
+- `{pm_enabled}` from `pm.enabled` (default: true)
+- `{pm_inbox_path}` — `pm-inbox.yaml` adjacent to `{epic_progress_path}`
+- `{fr_coverage_map_path}` — `fr-coverage-map.md` adjacent to `{epic_progress_path}`
+- `{backlog_path}` from `project.backlog_path`
+
+If `{pm_enabled}` is false, respond: "PM agent is disabled. Enable it in `pipeline-config.yaml` under `pm.enabled`."
+
+### Step 2: Classify Message
+
+Classify the user's message:
+
+| Type | Pattern | Example |
+|------|---------|---------|
+| `feature-request` | New functionality or scope expansion | "I want to add keyboard shortcuts" |
+| `priority-override` | Explicit reprioritization of existing stories | "Move KANBAN-015 above KANBAN-010" |
+| `modification` | Change to an existing requirement or story | "CSV export should also support JSON" |
+| `question` | Query about PRD state, coverage, or priorities | "What's the current PRD coverage?" |
+
+### Step 3: Append to Inbox
+
+Read `{pm_inbox_path}`. If the file does not exist, create it:
+
+```yaml
+messages: []
+```
+
+Determine the next sequential ID (max existing ID + 1, or 1 if empty).
+
+Append:
+
+```yaml
+  - id: {next_id}
+    timestamp: {ISO-8601}
+    from: user
+    type: {classified_type}
+    content: "{user's original message}"
+    status: pending
+```
+
+Write the updated file.
+
+### Step 4: Acknowledge
+
+Respond to the user:
+
+- **If a pipeline is actively running between sprints (PM is alive):** "Message queued for PM. PM is active and will process it shortly."
+- **If a pipeline is running but in execution (PM is not alive):** "Message #{id} queued for PM. PM will process it at the start of the next sprint cycle."
+- **If no pipeline is running:** "Message #{id} queued for PM. PM will process it when the next epic or project run starts."
+
+### Step 5: Handle Special Cases
+
+#### Priority Overrides — Apply Immediately
+
+For `priority-override` messages, also apply the change directly to `{backlog_path}`:
+
+1. Read `{backlog_path}`
+2. Identify the referenced stories
+3. If both stories exist, swap or reorder their `priority` fields
+4. Renumber surrounding priorities to maintain a gap-free sequence
+5. Write the updated backlog
+6. Update the inbox message:
+
+```yaml
+  - id: {id}
+    timestamp: {ISO-8601}
+    from: user
+    type: priority-override
+    content: "{original message}"
+    status: applied
+    applied_changes:
+      - story: {STORY-ID}, old_priority: {n}, new_priority: {n}
+      - story: {STORY-ID}, old_priority: {n}, new_priority: {n}
+```
+
+7. Confirm to user: "Priority override applied. {STORY-ID} moved from priority {old} to {new}. PM will reconcile the FR coverage map next sprint."
+
+If the referenced stories don't exist in the backlog, keep `status: pending` and respond: "Stories not found in backlog. PM will review this message when it next spawns."
+
+#### Questions — Answer from Disk
+
+For `question` messages, provide an immediate answer if possible:
+
+1. Read `{fr_coverage_map_path}` (if it exists)
+2. Read `{backlog_path}`
+3. Answer the question from current state
+4. Still queue the message so PM has a record
+
+If `{fr_coverage_map_path}` doesn't exist (no sprint has run yet), respond: "No FR coverage data available yet. PM will build the coverage map when the first sprint starts."
+
+## Notes
+
+- This skill can be invoked at any time — before, during, or between pipeline runs.
+- Messages persist in `pm-inbox.yaml` across conversation resets.
+- PM processes all pending messages at the start of each inter-sprint cycle.
+- Priority overrides are the only message type that takes immediate effect.

package/skills/valent-run-epic/SKILL.md

@@ -68,6 +68,17 @@ last_updated: {ISO-8601}
 ## Story Outcomes (compact — max 2 lines per story)
 ```
 
+### Step 3b: Initial PM Review
+
+If `{pm_enabled}` is true:
+
+1. Read and follow `.valent-pipeline/steps/orchestration/sprint-pm-review.md` with `{sprint_number}` = 1.
+2. PM builds the initial `fr-coverage-map.md` by reading the PRD and mapping FRs to backlog stories.
+3. PM processes any pre-run messages in `{pm_inbox_path}` (user may have queued ideas before starting the epic).
+4. PM validates backlog priority ordering against product value and may reprioritize.
+5. PM sends `[BACKLOG-UPDATE]` — re-read `{backlog_path}` after PM's update (PM may have added gap stories or reordered priorities).
+6. PM stays alive into Step 4d (grooming) for FR-tag oversight.
+
 ### Step 4: Sprint Loop
 
 Loop sprints until all epic stories are shipped, blocked, or cancelled. Each sprint follows the cycle: **groom → size → plan → execute → review → retro**.
@@ -103,7 +114,7 @@ BEND/FEND estimate Fibonacci points per story (filtered by `testing_profiles`).
 
 Read and follow `.valent-pipeline/steps/orchestration/sprint-plan.md`.
 
-Pack stories into sprint capacity by priority. Write `sprint-{n}-plan.md` and `sprint-{n}-status.yaml`. Kill Phase 1 agents (except Knowledge).
+Pack stories into sprint capacity by priority. Write `sprint-{n}-plan.md` and `sprint-{n}-status.yaml`. Kill Phase 1 agents (except Knowledge). PM teardown occurs during sprint-groom.md Step 6b (before sizing).
 
 #### 4g. Sprint Execution
 
@@ -142,6 +153,14 @@ Context compression may fire here — that is expected. The PostCompact hook re-
 
 When all stories in the epic are shipped (or remaining are all blocked/cancelled):
 
+#### 5 (pre). PM Final Audit
+
+If `{pm_enabled}` is true:
+
+Read and follow `.valent-pipeline/steps/orchestration/sprint-pm-audit.md`.
+
+PM spawns one final time, performs full FR coverage reconciliation, resolves weak-coverage items (these are blockers — "shipped but unproven" is not acceptable), writes `prd-audit-report.md`, and tears down. Include PM's audit summary in the epic report below.
+
 #### 5a. Write Epic Report
 
 Write `epic-report.md` adjacent to `{epic_progress_path}`:

package/skills/valent-run-project/SKILL.md

@@ -79,6 +79,17 @@ last_updated: {ISO-8601}
 ## Story Outcomes (compact — max 2 lines per story)
 ```
 
+### Step 3b: Initial PM Review
+
+If `{pm_enabled}` is true:
+
+1. Read and follow `.valent-pipeline/steps/orchestration/sprint-pm-review.md` with `{sprint_number}` = 1 and `{is_project_mode}` = true.
+2. PM reads ALL PRDs across all epics and builds a unified cross-epic FR coverage map. Each FR is tagged with its source epic so gaps route to the correct epic for story generation.
+3. PM processes any pre-run messages in `{pm_inbox_path}`.
+4. PM validates backlog priority ordering against product value across all epics and may reprioritize.
+5. PM sends `[BACKLOG-UPDATE]` — re-read `{backlog_path}` after PM's update.
+6. PM stays alive into Step 4d (grooming) for FR-tag oversight.
+
 ### Step 4: Sprint Loop
 
 Loop sprints until all stories are shipped, blocked, or cancelled. Each sprint follows the cycle: **groom → size → plan → execute → review → retro**.
@@ -117,7 +128,7 @@ BEND/FEND estimate Fibonacci points per story (filtered by `testing_profiles`).
 
 Read and follow `.valent-pipeline/steps/orchestration/sprint-plan.md`.
 
-Pack stories into sprint capacity by priority (cross-epic). Write `sprint-{n}-plan.md` and `sprint-{n}-status.yaml`. Kill Phase 1 agents (except Knowledge).
+Pack stories into sprint capacity by priority (cross-epic). Write `sprint-{n}-plan.md` and `sprint-{n}-status.yaml`. Kill Phase 1 agents (except Knowledge). PM teardown occurs during sprint-groom.md Step 6b (before sizing).
 
 #### 4g. Sprint Execution
 
@@ -158,6 +169,14 @@ Context compression may fire here — that is expected. The PostCompact hook re-
 
 When all stories are shipped (or remaining are all blocked/cancelled):
 
+#### 5 (pre). PM Final Audit
+
+If `{pm_enabled}` is true:
+
+Read and follow `.valent-pipeline/steps/orchestration/sprint-pm-audit.md`.
+
+PM spawns one final time with `{is_project_mode}` = true, reads all PRDs, performs full cross-epic FR coverage reconciliation, resolves weak-coverage items (blockers), writes `prd-audit-report.md`, and tears down. Include PM's audit summary in the project report below.
+
 #### 5a. Write Project Report
 
 Write `project-report.md` adjacent to `{epic_progress_path}`:

package/src/lib/config-schema.js

@@ -82,6 +82,19 @@ export function validateConfig(config) {
     }
   }
 
+  // PM section (optional)
+  if (config.pm) {
+    if (config.pm.enabled !== undefined && typeof config.pm.enabled !== 'boolean') {
+      errors.push(`pm.enabled must be a boolean, got: ${typeof config.pm.enabled}`);
+    }
+    if (config.pm.auto_reprioritize !== undefined && typeof config.pm.auto_reprioritize !== 'boolean') {
+      errors.push(`pm.auto_reprioritize must be a boolean, got: ${typeof config.pm.auto_reprioritize}`);
+    }
+    if (config.pm.auto_generate_gap_stories !== undefined && typeof config.pm.auto_generate_gap_stories !== 'boolean') {
+      errors.push(`pm.auto_generate_gap_stories must be a boolean, got: ${typeof config.pm.auto_generate_gap_stories}`);
+    }
+  }
+
   if (config.knowledge?.mode === 'sqlite' && !config.knowledge?.sqlite_db_path) {
     errors.push('knowledge.sqlite_db_path is required when knowledge.mode is "sqlite"');
   }
@@ -115,7 +128,7 @@ export const defaults = {
     state_management: 'React Context',
   },
   models: {
-    opus: ['BEND', 'FEND', 'CRITIC'],
+    opus: ['BEND', 'FEND', 'CRITIC', 'PM'],
     sonnet: ['REQS', 'UXA', 'QA-A', 'QA-B', 'READINESS', 'JUDGE', 'PMCP', 'Retrospective', 'DATA', 'MCP-DEV', 'LIBDEV', 'DOCGEN', 'IAC'],
     haiku: ['Knowledge', 'Embed', 'Help'],
   },
@@ -150,4 +163,9 @@ export const defaults = {
     recommended_context_window: '200k',
     epic_progress_path: './epic-progress.md',
   },
+  pm: {
+    enabled: true,
+    auto_reprioritize: true,
+    auto_generate_gap_stories: false,
+  },
 };