@pieerry/harness-kit 3.1.2 → 3.3.0

package/.claude/.hk-version

@@ -0,0 +1 @@
+3.1.2

package/.claude/agents/product-manager.md

@@ -12,9 +12,9 @@ When invoked, run /product-manager:run end to end. Follow .claude/plugins/produc
 Ask once if missing: team or squad, problem in 1-2 sentences, customers, hypothesis, bet link.
 
 Operating rules:
-- English by default. Domain terms stay native if the team uses them.
+- English by default. Domain terms stay native if team uses them.
 - Never invent. Mark gaps with `NOT FOUND - NEEDS REVIEW: {detail}`.
 - Voice: read .claude/plugins/product-manager/guides/writing-style.md. No em-dashes. Mermaid not ASCII.
-- Specific over generic. Real numbers, real names, real quotes.
+- Specific over generic. Real numbers, names, quotes.
 
 Return format: see .claude/commands/product-manager/run.md. Include paths, scores, attempts, publish status, blockers.

package/.claude/agents/staff-software-engineer.md

@@ -15,8 +15,8 @@ Ask once if missing:
 - area: backend, web, mobile, or devops (auto-detect from repo files if possible)
 
 Operating rules:
-- English by default. Domain terms stay native if the team uses them.
-- Read project conventions first. The repo's .claude/conventions/{area}.md (if present) overrides this plugin's defaults. See .claude/plugins/staff-software-engineer/guides/conventions-override.md.
+- English by default. Domain terms stay native if team uses them.
+- Read project conventions first. Repo's .claude/conventions/{area}.md (if present) overrides plugin defaults. See .claude/plugins/staff-software-engineer/guides/conventions-override.md.
 - Match repo style. Read 3+ similar files before writing.
 - Never invent class names, file paths, helpers. If unknown, `TBD - verify with tech lead`.
 - Voice: read .claude/plugins/staff-software-engineer/guides/coding-style.md and .claude/plugins/staff-software-engineer/guides/commit-style.md. No em-dashes. Mermaid not ASCII.

package/.claude/commands/pipeline/continue.md

@@ -2,14 +2,14 @@
 description: Resume the active pipeline at the next pending stage.
 ---
 
-Resume the active pipeline.
+Resume active pipeline.
 
-1. Read state: `python3 .claude/scripts/pipeline.py read`. If no `feature_id` or `current`, tell the user the pipeline is idle and suggest `/product-manager:run` or `/sse:run`.
-2. Read the next command: `python3 .claude/scripts/pipeline.py next`. This prints the slash command to invoke.
-3. Show the current status line: `python3 .claude/scripts/pipeline.py render`.
-4. Invoke the next-stage command. Use the recorded `feature_id` so artifacts land under the same name.
-5. After the stage completes (file written, approval marker applied), the post-write/post-edit hooks update state automatically. Run `/pipeline:continue` again to chain into the next stage, or stop here.
+1. Read state: `python3 .claude/scripts/pipeline.py read`. No `feature_id` or `current`, tell user pipeline idle. Suggest `/product-manager:run` or `/sse:run`.
+2. Read next command: `python3 .claude/scripts/pipeline.py next`. Prints slash command to invoke.
+3. Show current status line: `python3 .claude/scripts/pipeline.py render`.
+4. Invoke next-stage command. Use recorded `feature_id` so artifacts land under same name.
+5. After stage completes (file written, approval marker applied), post-write/post-edit hooks update state. Run `/pipeline:continue` again to chain next stage, or stop.
 
-If state has `pipeline` but no `feature_id` yet (intent recorded, first stage not started), invoke the first stage in `pipeline`.
+If state has `pipeline` but no `feature_id` yet (intent recorded, first stage not started), invoke first stage in `pipeline`.
 
-If the user wants to abandon the run instead, suggest `/pipeline:reset`.
+User wants to abandon run, suggest `/pipeline:reset`.

package/.claude/commands/pipeline/reset.md

@@ -2,10 +2,10 @@
 description: Clear the pipeline state file. Abandons the active feature run.
 ---
 
-Reset the pipeline.
+Reset pipeline.
 
-1. Confirm with the user that they want to abandon the active pipeline. Show `python3 .claude/scripts/pipeline.py render` so they see what they are dropping.
+1. Confirm user wants to abandon active pipeline. Show `python3 .claude/scripts/pipeline.py render` so they see what they drop.
 2. On confirmation: `python3 .claude/scripts/pipeline.py clear`.
-3. Output files under `.claude/plugins/*/outputs/` are not deleted. Only the in-memory state is cleared.
+3. Output files under `.claude/plugins/*/outputs/` not deleted. Only in-memory state cleared.
 
-The status bar will return to idle after reset.
+Status bar returns to idle after reset.

package/.claude/commands/product-manager/prd.md

@@ -2,13 +2,13 @@
 description: Generate a Product Requirements Document for an team squad. Business-facing artifact. Sensors and evals gate.
 ---
 
-Generate a PRD. Follow .claude/plugins/product-manager/guides/pipeline.md for retry, approval, and publish.
+Generate PRD. Follow .claude/plugins/product-manager/guides/pipeline.md for retry, approval, publish.
 
-Print a header card before drafting and a footer card after gates run. Format: .claude/scripts/stage-card.md.
+Print header card before drafting and footer card after gates run. Format: .claude/scripts/stage-card.md.
 
 Ask once if missing: squad, problem in 1-2 sentences, customers, hypothesis, bet link, stage.
 
-Compute feature_id = {YYYY-MM-DD}-{squad}-{slug}. Before generating, write the phase start marker:
+Compute feature_id = {YYYY-MM-DD}-{squad}-{slug}. Before generating, write phase start marker:
 
 ```
 .claude/plugins/product-manager/outputs/.markers/{feature_id}.prd-generate.start
@@ -30,7 +30,7 @@ Sensors: .claude/plugins/product-manager/sensors/prd-structure.md, .claude/plugi
 
 Evals: .claude/plugins/product-manager/evals/prd-quality.md, .claude/plugins/product-manager/evals/prd-readiness.md.
 
-After save, reply with this exact shape (name the actual sensors/evals/guides that ran, do not abbreviate):
+After save, reply with this exact shape (name actual sensors/evals/guides that ran, don't abbreviate):
 
 ```
 PRD saved at {path}.

package/.claude/commands/product-manager/prp.md

@@ -2,15 +2,15 @@
 description: Generate a Product Requirements Prompt for engineering handoff. Needs an approved PRD. Sensors, link validation, and eval gates.
 ---
 
-Generate a PRP. Follow .claude/plugins/product-manager/guides/pipeline.md for retry, approval, and publish.
+Generate PRP. Follow .claude/plugins/product-manager/guides/pipeline.md for retry, approval, publish.
 
-Print a header card before drafting and a footer card after gates run. Format: .claude/scripts/stage-card.md.
+Print header card before drafting and footer card after gates run. Format: .claude/scripts/stage-card.md.
 
-Source PRD: if user passes a path, use it. Else pick the most recent in .claude/plugins/product-manager/outputs/prd/. None found, abort. Tell user to run /product-manager:prd first. .claude/plugins/product-manager/hooks/pre-prp-check.sh blocks if the PRD lacks the approved marker.
+Source PRD: user passes path, use it. Else pick most recent in .claude/plugins/product-manager/outputs/prd/. None found, abort. Tell user to run /product-manager:prd first. .claude/plugins/product-manager/hooks/pre-prp-check.sh blocks if PRD lacks approved marker.
 
-Compute feature_id from the source PRD filename (basename without .md). Save the PRP to .claude/plugins/product-manager/outputs/prp/{feature_id}.md so it matches.
+Compute feature_id from source PRD filename (basename without .md). Save PRP to .claude/plugins/product-manager/outputs/prp/{feature_id}.md so it matches.
 
-Before generating, write the phase start marker:
+Before generating, write phase start marker:
 
 ```
 .claude/plugins/product-manager/outputs/.markers/{feature_id}.prp-generate.start
@@ -19,7 +19,7 @@ Before generating, write the phase start marker:
 Content: `{"timestamp": "<ISO-8601 UTC now>", "session_id": ""}`
 
 Read:
-- the source PRD
+- source PRD
 - .claude/plugins/product-manager/guides/prp-guidelines.md
 - .claude/plugins/product-manager/guides/writing-style.md
 - .claude/plugins/product-manager/guides/templates/prp.md
@@ -34,7 +34,7 @@ Sensors: .claude/plugins/product-manager/sensors/prp-structure.md, .claude/plugi
 
 Evals: .claude/plugins/product-manager/evals/prp-quality.md, .claude/plugins/product-manager/evals/prp-context-readiness.md.
 
-After save, reply with this exact shape (name the actual sensors/evals/guides that ran):
+After save, reply with this exact shape (name actual sensors/evals/guides that ran):
 
 ```
 PRP saved at {path}.

package/.claude/commands/product-manager/run.md

@@ -4,14 +4,14 @@ description: Run the full PM pipeline. PRD then PRP, with sensor and eval gates
 
 Run end to end.
 
-1. Invoke /product-manager:prd. Wait for the approval marker.
-2. .claude/plugins/product-manager/hooks/pre-prp-check.sh validates the PRD.
-3. Invoke /product-manager:prp using the just-approved PRD.
+1. Invoke /product-manager:prd. Wait for approval marker.
+2. .claude/plugins/product-manager/hooks/pre-prp-check.sh validates PRD.
+3. Invoke /product-manager:prp using just-approved PRD.
 4. Return summary.
 
 Follow .claude/plugins/product-manager/guides/pipeline.md. Read .claude/agents/product-manager.md for inputs and rules.
 
-Return format. Name every sensor, eval, and guide that ran. Generic summaries are not acceptable — list specifics so the user sees what was checked and what was loaded.
+Return format. Name every sensor, eval, guide that ran. Generic summaries not acceptable — list specifics so user sees what was checked and loaded.
 
 ```
 Pipeline complete.
@@ -34,4 +34,4 @@ Blockers:
 - {file:line, issue, fix}
 ```
 
-If a phase has no sensors/eval/guides/refs, omit that line rather than printing an empty one.
+Phase has no sensors/eval/guides/refs, omit line rather than print empty one.

package/.claude/commands/sse/dev.md

@@ -2,37 +2,37 @@
 description: Implement the approved plan in code. Writes commits, runs gates, returns diff summary.
 ---
 
-Implement the plan. Follow .claude/plugins/staff-software-engineer/guides/pipeline.md.
+Implement plan. Follow .claude/plugins/staff-software-engineer/guides/pipeline.md.
 
-Print a header card before coding and a footer card after gates run. Format: .claude/scripts/stage-card.md.
+Print header card before coding and footer card after gates run. Format: .claude/scripts/stage-card.md.
 
-Source plan: latest in .claude/plugins/staff-software-engineer/outputs/plan/ with approved marker. If none, abort and ask user to run /sse:plan first.
+Source plan: latest in .claude/plugins/staff-software-engineer/outputs/plan/ with approved marker. None, abort. Ask user to run /sse:plan first.
 
-Before coding, write the phase start marker:
+Before coding, write phase start marker:
 
 ```
 .claude/plugins/staff-software-engineer/outputs/.markers/{feature_id}.dev.start
 ```
 
 Read:
-- the source plan
-- the area skill: .claude/plugins/staff-software-engineer/skills/{area}/SKILL.md
+- source plan
+- area skill: .claude/plugins/staff-software-engineer/skills/{area}/SKILL.md
 - .claude/plugins/staff-software-engineer/guides/coding-style.md
 - .claude/plugins/staff-software-engineer/guides/commit-style.md
 - project conventions: {repo}/.claude/conventions/{area}.md if present
 
-Read 3+ similar files in the target repo before writing. Match conventions:
+Read 3+ similar files in target repo before writing. Match conventions:
 - framework version (Spring 4.3 no-Boot vs Spring Boot, Vue 2 vs 3)
 - build tool (Maven, Gradle, npm)
 - package layout, test framework, helpers
 
 Write code in small commits (1-4 files, < 100 lines ideal). Conventional Commits format.
 
-Gates (run after each implementation step):
+Code gates (run after each implementation step):
 - .claude/plugins/staff-software-engineer/sensors/code-conventions.md (lint, formatting, banned patterns)
 - .claude/plugins/staff-software-engineer/sensors/test-coverage.md (every feature/bugfix has tests)
 
-If any gate fails, fix and retry. Max 3 attempts. Hard stop after 3.
+Any gate fails, fix and retry. Max 3 attempts. Hard stop after 3.
 
 After done, write `.claude/plugins/staff-software-engineer/outputs/dev/{feature_id}.md` with summary:
 - files changed
@@ -40,19 +40,24 @@ After done, write `.claude/plugins/staff-software-engineer/outputs/dev/{feature_
 - gate results
 - blockers if any
 
-Append approval marker when all gates pass:
+Document gates (run on saved summary):
+- Sensor: .claude/plugins/staff-software-engineer/sensors/dev-structure.md (auto-run by post-write hook)
+- Eval:   .claude/plugins/staff-software-engineer/evals/dev-quality.md (you score it; threshold 8.0)
+
+Append approval marker only when code gates pass and dev-quality eval is >= 8.0:
 
 ```
 <!-- approved: {YYYY-MM-DD} -->
 ```
 
-After approval, reply with this exact shape (name the actual sensors/guides that ran):
+After approval, reply with this exact shape (name actual sensors/evals/guides that ran):
 
 ```
 Dev complete. branch {branch}.
   files changed: {N}
   commits: {M} ({short-sha}, {short-sha}, ...)
-  sensors: code-conventions ok, test-coverage ok
+  sensors: code-conventions ok, test-coverage ok, dev-structure ok
+  eval:    dev-quality {N}/10
   guides:  coding-style.md, commit-style.md, skills/{area}/SKILL.md
   refs:    plan/{feature_id}.md, conventions/{area}.md
   next:    /sse:test

package/.claude/commands/sse/plan.md

@@ -2,15 +2,15 @@
 description: Generate an implementation plan from an approved PRP. Sensors and evals gate.
 ---
 
-Generate a technical plan. Follow .claude/plugins/staff-software-engineer/guides/pipeline.md for retry, approval, and publish.
+Generate technical plan. Follow .claude/plugins/staff-software-engineer/guides/pipeline.md for retry, approval, publish.
 
-Print a header card before drafting and a footer card after gates run. Format: .claude/scripts/stage-card.md.
+Print header card before drafting and footer card after gates run. Format: .claude/scripts/stage-card.md.
 
-Source PRP: if user passes a path, use it. Else pick the most recent in .claude/plugins/product-manager/outputs/prp/. None found, abort. Tell user to run /product-manager:prp first.
+Source PRP: user passes path, use it. Else pick most recent in .claude/plugins/product-manager/outputs/prp/. None found, abort. Tell user to run /product-manager:prp first.
 
-Compute feature_id from the source PRP filename (basename without .md). Save the plan to .claude/plugins/staff-software-engineer/outputs/plan/{feature_id}.md so it matches.
+Compute feature_id from source PRP filename (basename without .md). Save plan to .claude/plugins/staff-software-engineer/outputs/plan/{feature_id}.md so it matches.
 
-Before generating, write the phase start marker:
+Before generating, write phase start marker:
 
 ```
 .claude/plugins/staff-software-engineer/outputs/.markers/{feature_id}.plan-generate.start
@@ -19,11 +19,11 @@ Before generating, write the phase start marker:
 Content: `{"timestamp": "<ISO-8601 UTC now>", "session_id": ""}`
 
 Read:
-- the source PRP
+- source PRP
 - .claude/plugins/staff-software-engineer/guides/pipeline.md
 - .claude/plugins/staff-software-engineer/guides/coding-style.md
 - .claude/plugins/staff-software-engineer/guides/examples/good-plan-example.md
-- the area-specific skill: .claude/plugins/staff-software-engineer/skills/{area}/SKILL.md (where area = backend, web, mobile, devops)
+- area-specific skill: .claude/plugins/staff-software-engineer/skills/{area}/SKILL.md (area = backend, web, mobile, devops)
 - project conventions if present: {repo}/.claude/conventions/{area}.md (see .claude/plugins/staff-software-engineer/guides/conventions-override.md)
 
 Save to .claude/plugins/staff-software-engineer/outputs/plan/{feature_id}.md.
@@ -32,7 +32,7 @@ Sensors: .claude/plugins/staff-software-engineer/sensors/plan-structure.md.
 
 Evals: .claude/plugins/staff-software-engineer/evals/plan-quality.md.
 
-After save, reply with this exact shape (name the actual sensors/evals/guides that ran):
+After save, reply with this exact shape (name actual sensors/evals/guides that ran):
 
 ```
 Plan saved at {path}.

package/.claude/commands/sse/pr-monitor.md

@@ -0,0 +1,56 @@
+---
+description: Watch the active PR for merge with backoff polling. Auto-invoked after /sse:pr. Clears pipeline state on merge.
+---
+
+Monitor active PR until it merges. Polls with backoff: 3min → 6min → 12min → 24min → 30min cap, escalating after 5 attempts at each rung.
+
+Runs in active session via ScheduleWakeup. Closing session ends monitor.
+
+## Flow
+
+1. **Read state** from `.claude/.pr-monitor-state.json`.
+   - Missing → first run. Detect PR from current branch:
+     ```
+     gh pr view --json number,url,headRefName,state
+     ```
+     No PR or state already MERGED → exit silently (nothing to watch).
+     Otherwise initialize state:
+     ```
+     .claude/scripts/pr-monitor.py init <number> <url> <branch>
+     ```
+     Script prints next-delay in seconds.
+   - Present → use `state.pr_number`.
+
+2. **Check merge status**:
+   ```
+   gh pr view <pr_number> --json state -q .state
+   ```
+
+3. **If `MERGED`**:
+   - Print one line: `PR #<n> merged → <url>. Pipeline state cleared. Start next: /product-manager:run`
+   - Clear monitor state: `.claude/scripts/pr-monitor.py clear`
+   - Clear pipeline state: `.claude/scripts/pipeline.py clear`
+   - **Do not** call ScheduleWakeup. Done.
+
+4. **If `CLOSED` (not merged)**:
+   - Print: `PR #<n> closed without merge. Monitor stopped.`
+   - Clear monitor state only. Done.
+
+5. **If `OPEN`**:
+   - Bump attempt counter, get next interval:
+     ```
+     .claude/scripts/pr-monitor.py bump
+     ```
+     Script prints next-delay in seconds (e.g. `180`, `360`, `720`, `1440`, `1800`).
+   - Read state with `.claude/scripts/pr-monitor.py read` for `current_interval_min` and `total_attempts` to include in user-facing line.
+   - Print one line: `PR #<n> still open (attempt <total>, next check in <m>min).`
+   - Schedule next wake:
+     - Load `ScheduleWakeup` via ToolSearch: `select:ScheduleWakeup`
+     - Call with `delaySeconds=<next>`, `prompt="/sse:pr-monitor"`, `reason="watching PR #<n> for merge, interval <m>min"`.
+
+## First invocation (right after /sse:pr)
+
+No state file yet. Initialize then schedule first wake at 3min (180s). Print:
+```
+PR monitor armed for #<n>. First check in 3min, escalates to 30min cap.
+```

package/.claude/commands/sse/pr.md

@@ -2,15 +2,15 @@
 description: Open a Pull Request on GitHub following team conventions. Draft by default.
 ---
 
-Open a Pull Request.
+Open Pull Request.
 
-Print a header card before opening and a footer card after gh returns. Format: .claude/scripts/stage-card.md.
+Print header card before opening and footer card after gh returns. Format: .claude/scripts/stage-card.md.
 
 Prerequisites:
 - All previous gates passed (plan approved, dev approved, test approved).
 - Branch pushed to origin with current changes.
 
-Before opening, write the phase start marker:
+Before opening, write phase start marker:
 
 ```
 .claude/plugins/staff-software-engineer/outputs/.markers/{feature_id}.pr.start
@@ -19,9 +19,9 @@ Before opening, write the phase start marker:
 Read:
 - .claude/plugins/staff-software-engineer/guides/pr-template.md
 - .claude/plugins/staff-software-engineer/guides/commit-style.md
-- the latest plan and dev outputs (for the body content)
+- latest plan and dev outputs (for body content)
 
-Detect ticket id from branch name (e.g., `feat/PROJ-123-foo` -> `PROJ-123`). If branch has none, ask the user once. Never call Jira API.
+Detect ticket id from branch name (e.g., `feat/PROJ-123-foo` -> `PROJ-123`). Branch has none, ask user once. Never call Jira API.
 
 Compose:
 - Title: conventional commit prefix + short description (e.g., `feat(PROJ-123): add timezone-aware deadline check`)
@@ -35,8 +35,14 @@ Save .claude/plugins/staff-software-engineer/outputs/pr/{feature_id}.md with:
 - title
 - draft status
 - summary
+- test plan checklist
+- refs (plan + dev paths)
 
-Append approval marker:
+Document gates (run on saved record):
+- Sensor: .claude/plugins/staff-software-engineer/sensors/pr-structure.md (auto-run by post-write hook)
+- Eval:   .claude/plugins/staff-software-engineer/evals/pr-quality.md (you score it; threshold 8.0)
+
+Append approval marker only when sensor passes and pr-quality eval is >= 8.0:
 
 ```
 <!-- approved: {YYYY-MM-DD} ready-for-handoff: true -->
@@ -46,9 +52,13 @@ Reply with this exact shape:
 
 ```
 PR opened: {url}
-  title:  {title}
-  draft:  {yes|no}
-  guides: pr-template.md, commit-style.md
-  refs:   plan/{feature_id}.md, dev/{feature_id}.md
-  next:   request review (if draft, mark ready when checks pass)
+  title:   {title}
+  draft:   {yes|no}
+  sensors: pr-structure ok
+  eval:    pr-quality {N}/10
+  guides:  pr-template.md, commit-style.md
+  refs:    plan/{feature_id}.md, dev/{feature_id}.md
+  next:    request review (if draft, mark ready when checks pass)
 ```
+
+After replying, **auto-invoke `/sse:pr-monitor`** so session watches PR for merge with backoff polling. Skip if user passed `--no-monitor` or PR already MERGED.

package/.claude/commands/sse/run.md

@@ -4,15 +4,15 @@ description: Run the full engineering pipeline. Plan, dev, test, pr in sequence.
 
 Run end to end.
 
-1. Invoke /sse:plan. Wait for the approval marker on the plan.
-2. Invoke /sse:dev. Implements the plan in code.
-3. Invoke /sse:test. Runs the project test suite.
-4. Invoke /sse:pr. Opens the pull request.
+1. Invoke /sse:plan. Wait for approval marker on plan.
+2. Invoke /sse:dev. Implements plan in code.
+3. Invoke /sse:test. Runs project test suite.
+4. Invoke /sse:pr. Opens pull request.
 5. Return summary.
 
-Follow .claude/plugins/staff-software-engineer/guides/pipeline.md for retry, approval markers, token accounting, and publish behavior.
+Follow .claude/plugins/staff-software-engineer/guides/pipeline.md for retry, approval markers, token accounting, publish behavior.
 
-Return format. Name every sensor, eval, and guide that ran. Generic summaries are not acceptable — list specifics so the user sees what was checked and what was loaded.
+Return format. Name every sensor, eval, guide that ran. Generic summaries not acceptable — list specifics so user sees what was checked and loaded.
 
 ```
 Engineering pipeline complete.
@@ -47,4 +47,4 @@ Blockers:
 - {file:line, issue, fix}
 ```
 
-If a phase has no sensors/eval/guides/refs, omit that line for that phase rather than printing an empty one.
+Phase has no sensors/eval/guides/refs, omit line for that phase rather than print empty one.

package/.claude/commands/sse/test.md

@@ -2,25 +2,25 @@
 description: Run the project test suite. Reports results to .claude/plugins/staff-software-engineer/outputs/test/.
 ---
 
-Run the test suite for the current repo.
+Run test suite for current repo.
 
-Print a header card before running and a footer card after the suite finishes. Format: .claude/scripts/stage-card.md.
+Print header card before running and footer card after suite finishes. Format: .claude/scripts/stage-card.md.
 
-Detect the project test command (in order):
+Detect project test command (in order):
 1. Check README.md or CONTRIBUTING.md for explicit test instructions.
 2. Maven (pom.xml present): `./mvnw test` or `mvn test`
 3. Gradle (build.gradle present): `./gradlew test`
 4. npm (package.json with "test" script): `npm test`
 5. pytest (pyproject.toml or pytest.ini): `pytest`
-6. Other: ask the user.
+6. Other: ask user.
 
-Before running, write the phase start marker:
+Before running, write phase start marker:
 
 ```
 .claude/plugins/staff-software-engineer/outputs/.markers/{feature_id}.test.start
 ```
 
-Run the test command. Capture stdout and stderr.
+Run test command. Capture stdout and stderr.
 
 Save .claude/plugins/staff-software-engineer/outputs/test/{feature_id}.md with:
 - command run
@@ -29,13 +29,17 @@ Save .claude/plugins/staff-software-engineer/outputs/test/{feature_id}.md with:
 - failing test names (if any)
 - duration
 
-Append approval marker when exit code is 0:
+Document gates (run on saved report):
+- Sensor: .claude/plugins/staff-software-engineer/sensors/test-structure.md (auto-run by post-write hook)
+- Eval:   .claude/plugins/staff-software-engineer/evals/test-quality.md (you score it; threshold 8.0)
+
+Append approval marker only when exit code is 0 and test-quality eval is >= 8.0:
 
 ```
 <!-- approved: {YYYY-MM-DD} -->
 ```
 
-If tests fail, return a blocker with the failing test names and a snippet of the failure output. Do not retry automatically; let the user decide.
+Tests fail, return blocker with failing test names and snippet of failure output. Don't retry automatically; let user decide.
 
 Reply with this exact shape:
 
@@ -45,8 +49,10 @@ Tests {passed|failed}.
   passed:   {N}
   failed:   {M}
   duration: {seconds}s
+  sensors:  test-structure ok
+  eval:     test-quality {N}/10
   output:   {path/to/test/output.md}
   next:     /sse:pr (if passed) | fix failing tests (if failed)
 ```
 
-If failed, append a `failures:` block listing each failing test name with a one-line snippet from the failure output.
+Failed, append `failures:` block listing each failing test name with one-line snippet from failure output.

package/.claude/conventions/README.md

@@ -0,0 +1,12 @@
+# Project Conventions
+
+Override staff-software-engineer plugin defaults for this repo. Create files as needed:
+
+- backend.md
+- web.md
+- mobile.md
+- devops.md
+
+When a file exists here, plugin reads it on top of defaults. Rules in this folder win.
+
+See `.claude/plugins/staff-software-engineer/guides/conventions-override.md` for the full reference.

package/.claude/hooks/activity-pre-read.sh

@@ -0,0 +1,18 @@
+#!/bin/sh
+# PreToolUse hook on Read. If the file being read is a sensor/eval/guide
+# under any plugin, mark current activity so status-line.sh can surface it.
+
+FILE_PATH="${CLAUDE_TOOL_FILE_PATH:-}"
+[ -z "$FILE_PATH" ] && exit 0
+
+KIND=""
+case "$FILE_PATH" in
+  */plugins/*/sensors/*.md) KIND=sensor ;;
+  */plugins/*/evals/*.md)   KIND=eval ;;
+  */plugins/*/guides/*.md)  KIND=guide ;;
+  *) exit 0 ;;
+esac
+
+NAME="$(basename "$FILE_PATH" .md)"
+[ -x ".claude/scripts/activity.py" ] && python3 .claude/scripts/activity.py set "$KIND" "$NAME" 2>/dev/null
+exit 0

package/.claude/hooks/pipeline-session-start.sh

@@ -1,6 +1,9 @@
 #!/bin/sh
-# SessionStart hook. If a pipeline state exists with an incomplete pipeline,
-# print a resume hint that the user can act on.
+# SessionStart hook.
+# 1. If current branch has a MERGED/CLOSED PR and pipeline state looks stale
+#    (no feature_id), clear state — the work shipped, nothing to resume.
+# 2. Otherwise, if a pipeline state exists with an incomplete pipeline,
+#    print a resume hint.
 
 set -e
 
@@ -9,6 +12,27 @@ STATE_FILE=".claude/.pipeline-state.json"
 [ -x "$PIPELINE_PY" ] || exit 0
 [ -f "$STATE_FILE" ] || exit 0
 
+# Stale-state auto-clear: branch has terminal PR + no feature_id recorded.
+if command -v gh >/dev/null 2>&1 && command -v python3 >/dev/null 2>&1; then
+  FID="$(python3 -c 'import json,sys
+try:
+    s=json.load(open(".claude/.pipeline-state.json"))
+    print(s.get("feature_id") or "")
+except Exception:
+    pass' 2>/dev/null || true)"
+  if [ -z "$FID" ]; then
+    PR_STATE="$(gh pr view --json state -q .state 2>/dev/null || true)"
+    PR_NUM="$(gh pr view --json number -q .number 2>/dev/null || true)"
+    case "$PR_STATE" in
+      MERGED|CLOSED)
+        python3 "$PIPELINE_PY" clear >/dev/null 2>&1 || true
+        printf 'previous feature shipped (PR #%s %s). pipeline state cleared.\nstart next with /product-manager:run or /sse:run\n' "$PR_NUM" "$PR_STATE"
+        exit 0
+        ;;
+    esac
+  fi
+fi
+
 CURRENT="$(python3 "$PIPELINE_PY" next 2>/dev/null || true)"
 [ -z "$CURRENT" ] && exit 0
 

package/.claude/hooks/status-line.sh

@@ -1,14 +1,29 @@
 #!/bin/sh
 # Status line for the harness-kit pipeline. Reads .claude/.pipeline-state.json
 # (managed by .claude/scripts/pipeline.py) for live state. Falls back to a
-# file-scan if state is absent.
+# file-scan if state is absent. Appends current activity (sensor/eval/guide)
+# in cyan when one is active.
 
 PIPELINE_PY=".claude/scripts/pipeline.py"
+ACTIVITY_PY=".claude/scripts/activity.py"
+
+CYAN=$(printf '\033[36m')
+RESET=$(printf '\033[0m')
+
+append_activity() {
+  if [ -x "$ACTIVITY_PY" ]; then
+    A="$(python3 "$ACTIVITY_PY" read 2>/dev/null || true)"
+    if [ -n "$A" ]; then
+      printf ' · %s%s%s' "$CYAN" "$A" "$RESET"
+    fi
+  fi
+}
 
 if [ -x "$PIPELINE_PY" ] && [ -f ".claude/.pipeline-state.json" ]; then
   OUT="$(python3 "$PIPELINE_PY" render 2>/dev/null)"
   if [ -n "$OUT" ]; then
     printf '%s' "$OUT"
+    append_activity
     exit 0
   fi
 fi
@@ -99,3 +114,4 @@ if [ -z "$OUT" ]; then
 else
   printf '%s' "$OUT"
 fi
+append_activity

package/.claude/plugins/product-manager/evals/prd-quality.md

@@ -9,21 +9,21 @@ Score each dimension 0-10. Cite line numbers when scoring below 7. Weighted tota
 ## Rubric
 
 ### Clarity (weight 20%)
-Is the problem stated in 1-2 sentences? Names who suffers, how often, what it costs?
+Problem stated in 1-2 sentences? Names who suffers, how often, what it costs?
 
 - 10: crisp, specific, evidence-backed
 - 5: present but generic
 - 0: missing or vague
 
 ### Hypothesis (weight 15%)
-Is there an "If we X, then Y will Z, because W" hypothesis with numeric target?
+"If we X, then Y will Z, because W" hypothesis with numeric target?
 
 - 10: falsifiable, numeric, evidence-tied
 - 5: directional but missing target or evidence
 - 0: aspirational, no measurable claim
 
 ### Customer specificity (weight 10%)
-Real customers named with reasons each one matters?
+Real customers named with reasons each matters?
 
 - 10: concrete, differentiated
 - 5: segments named but thin

package/.claude/plugins/product-manager/evals/prd-readiness.md

@@ -3,7 +3,7 @@
 Type: rule-based stage gate
 Mode: advisory
 
-Decides whether a PRD has enough content for its declared stage to advance. Run after prd-quality passes.
+Decides whether PRD has enough content for declared stage to advance. Run after prd-quality passes.
 
 ## Stage gates