prizmkit 1.1.154 → 1.1.156

package/bundled/dev-pipeline/templates/bugfix-bootstrap-prompt.md

@@ -41,24 +41,30 @@ You are the **bug fix session agent**. Fix Bug {{BUG_ID}}: "{{BUG_TITLE}}".
 {{GLOBAL_CONTEXT}}
 
 
-## ⚠️ Context Budget Rules (CRITICAL)
+## Context and Execution Rules
 
-0. **NON-INTERACTIVE MODE** — NEVER ask for confirmation. Proceed autonomously.
-1. **context-snapshot.md is your single source of truth** — After it is built, read context-snapshot.md instead of re-reading individual source files.
-2. **Never re-read your own writes** — Trust your write was correct.
-3. **Never re-read unmodified files** — Once you read a file into context, do NOT read it again unless it was modified by an Edit/Write after your last read. Use `grep -n` or `sed -n` for targeted lookups instead of full-file Read. Before issuing a Read, ask: "Have I read this file before, and has it changed since?" If no — do NOT read it. Follow the active skill and runtime contracts when an execution unit is created; do not infer its permissions, continuation, or retry behavior from this bootstrap.
-4. **Stay focused** — Do NOT explore code unrelated to this bug.
-4. **Minimize tool output** — Capture to temp file, scan head/tail, filter with grep/sed/awk. Never load full output.
-5. **No intermediate commits** — All changes committed once at the end via `/prizmkit-committer`.
+- Resolve safe uncertainty from the task contract, current code, and project docs; otherwise record a truthful blocker.
+- Use `context-snapshot.md` for navigation, not as a substitute for current source, diffs, or authoritative artifacts.
+- Avoid repeated full reads, but reread bounded changed/stale ranges and verify every modification.
+- Keep scope tied to the bug, concrete coupling, and mandatory gates. Bound command output with host-appropriate temporary locations.
+- `/prizmkit-test` exclusively owns the complete test and repair loop.
+- Do not stage, commit, amend, merge, push, reset, clean, or hand-edit runner-owned state.
 
-## Bug Fix Artifacts Directory
+## Bug Fix Artifact Manifest
 
 ```
 .prizmkit/bugfix/{{BUG_ID}}/
-├── spec.md              ← /prizmkit-plan output (root cause, scope, behavior preservation)
-├── plan.md              ← /prizmkit-plan output (fix tasks, first task = reproduction test)
-├── context-snapshot.md  ← Project context for this bug
-└── fix-report.md        ← Final bug resolution report
+├── spec.md
+├── plan.md
+├── context-snapshot.md
+├── review-report.md
+├── test-report.md
+├── test-result.json
+├── fix-report.md
+├── retrospective-result.json
+├── failure-log.md
+├── workflow-checkpoint.json
+└── runtime-commit-request.json
 ```
 
 {{CHECKPOINT_SYSTEM}}
@@ -116,13 +122,13 @@ Use this protocol only when the bug is UI/frontend-reproducible. Evidence must c
 
 Write `.prizmkit/bugfix/{{BUG_ID}}/context-snapshot.md`:
 - **Section 1 — Bug Brief**: bug description + acceptance criteria + root cause analysis
-- **Section 2 — Affected Files**: full verbatim content of each affected source file
-- **Section 3 — Existing Tests**: full content of related test files
+- **Section 2 — Affected Files**: paths, relevant interfaces, and why each file matters
+- **Section 3 — Existing Tests**: relevant test paths, behavior covered, and gaps; do not copy full files
 - **Section 4 — Prizm Context**: relevant TRAPS, RULES from .prizmkit/prizm-docs/
 
 **Step 3 — Plan the fix**:
 
-Run `/prizmkit-plan` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
+Invoke `/prizmkit-plan` once with the exact artifact root, bug requirement, and diagnosis context. It returns only `PLAN_READY` or `PLAN_BLOCKED`:
 - The spec.md should capture: root cause, impact scope, behavior that must be preserved
 - The plan.md Tasks section **MUST start with a reproduction test task** — a test that FAILS with current code (RED state), proving the bug exists
 - Subsequent tasks implement the minimal fix to make the test pass (GREEN state)
@@ -136,7 +142,7 @@ Run `/prizmkit-plan` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
 
 **CP-1**: spec.md and plan.md exist with Tasks section.
 
-**Checkpoint update**: Set step `bug-diagnosis-and-plan` to `"completed"`.
+**Checkpoint update**: This injected session sets `bug-diagnosis-and-plan` completed with `stage_result=PLAN_READY` only after artifact validation; `PLAN_BLOCKED` is recorded as failed.
 
 {{IF_VERIFICATION_MANUAL_OR_HYBRID}}
 **NOTE**: verification_type is '{{VERIFICATION_TYPE}}'. The plan.md MUST also include:
@@ -150,27 +156,20 @@ Run `/prizmkit-plan` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
 
 **Goal**: Execute the fix plan. Reproduction test goes from RED → GREEN.
 
-Run `/prizmkit-implement` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
-- Executes plan.md tasks in order (TDD: first task creates failing test, subsequent tasks fix the code)
-- Marks each task `[x]` on completion
-- Runs test suite after each task
-- Uses convergence-based test failure recovery (keep fixing while progress is being made)
-
-{{IF_BROWSER_INTERACTION}}
-
-**Browser Verification During Implementation**: If the bug is UI-reproducible, apply the Browser Verification Protocol above and document evidence in the implementation notes.
-
-{{END_IF_BROWSER_INTERACTION}}
+Invoke `/prizmkit-implement` with the exact artifact root, reviewed plan, and any explicit repair findings. It returns only `IMPLEMENTED` or `IMPLEMENT_BLOCKED`:
+- Execute plan.md tasks in order (TDD: first task creates a failing test, subsequent tasks fix the code)
+- Mark each task `[x]` on completion
+- Use only focused implementation checks; the authoritative test gate runs after review
 
 After implement completes, verify:
 1. All tasks in plan.md are `[x]`
 2. Reproduction test passes (GREEN)
-3. Full test suite passes (no regression)
-4. Each acceptance criterion is met
+3. Focused implementation checks pass
+4. Each acceptance criterion has implementation evidence
 
 **CP-2**: All tasks complete, reproduction test passes, no regression.
 
-**Checkpoint update**: Set step `prizmkit-implement` to `"completed"`.
+**Checkpoint update**: This injected session maps `IMPLEMENTED` to completed after validation and stops on `IMPLEMENT_BLOCKED`.
 
 ---
 
@@ -178,8 +177,8 @@ After implement completes, verify:
 
 Code review is mandatory for every committable Bugfix. Do not skip this phase for small or obvious fixes.
 
-Run `/prizmkit-code-review` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
-- The skill runs up to ten Main-Agent review rounds, directly repairs accepted findings, and verifies each repair
+Invoke `/prizmkit-code-review` with the exact artifact root, spec, plan, and complete current change. It returns only `PASS` or `NEEDS_FIXES`:
+- The Skill runs up to ten Main-Agent review rounds, directly repairs accepted findings, and verifies each repair
 - Missing required evidence, unsafe repair, failed verification, or round-ten non-convergence produces `NEEDS_FIXES`
 - The skill appends Main-Agent review and repair progress to `review-report.md`
 
@@ -187,7 +186,7 @@ Run `/prizmkit-code-review` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
 ```bash
 {{RUNTIME_HELPER_CMD}} text final-verdict .prizmkit/bugfix/{{BUG_ID}}/review-report.md
 ```
-Handle `REVIEW_PASS` and `REVIEW_NEEDS_FIXES` according to task policy. `REVIEW_INCOMPLETE` or `REVIEW_INVALID` means review did not complete: inspect progress once, write `failure-log.md`, and stop. Do not accept stale verdict text or replace required independent review with Main-Agent self-review.
+This injected session maps Skill `PASS`/`NEEDS_FIXES` to `REVIEW_PASS`/`REVIEW_NEEDS_FIXES`. For `NEEDS_FIXES`, classify concrete findings into an allowed `repair_scope`; safe in-budget repair re-enters Implement through the helper, while unknown/unsafe scope blocks. `REVIEW_INCOMPLETE` or `REVIEW_INVALID` means review did not complete: inspect progress once, write `failure-log.md`, and stop. Do not accept stale verdict text or replace required independent review with Main-Agent self-review.
 
 {{IF_BROWSER_INTERACTION}}
 
@@ -195,21 +194,33 @@ Handle `REVIEW_PASS` and `REVIEW_NEEDS_FIXES` according to task policy. `REVIEW_
 
 {{END_IF_BROWSER_INTERACTION}}
 
-**CP-3**: Code review complete, `review-report.md` written, and all tests green.
+**CP-3**: Code review returned `PASS`, `review-report.md` is valid, and review-local verification is recorded.
 
 **Checkpoint update**: Set step `prizmkit-code-review` to `"completed"` only for an accepted valid Final Result.
 
 ---
 
+{{IF_BROWSER_INTERACTION}}
+### Phase 4: Configured Browser Verification
+
+After valid code-review PASS and before Test, execute the configured Browser Verification Protocol above. Record evidence and cleanup. A blocking failure stops before Test; a non-blocking unavailable check is recorded and may continue according to task policy.
+
+{{END_IF_BROWSER_INTERACTION}}
+{{IF_VERIFICATION_MANUAL_OR_HYBRID}}
+### Phase 4: Manual / Hybrid Verification
+
+Complete the declared UAT checklist automatically with available browser, CLI, API, fixture, and test evidence. Record the evidence before Test. If required evidence cannot be established, write `failure-log.md` and stop truthfully.
+
+{{END_IF_VERIFICATION_MANUAL_OR_HYBRID}}
 ### Phase 4: Post-Review Test Gate — PrizmKit Test
 
-After a valid code-review PASS, run the mandatory test gate against the same artifact root:
+After valid code-review PASS and all configured Browser/Manual verification, run the mandatory test gate against the same artifact root:
 
 ```text
 /prizmkit-test scope=this-change artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/
 ```
 
-Read the sibling `test-report.md` and `test-result.json`; they must agree. `TEST_PASS` proceeds to retrospective and commit. `TEST_NEEDS_FIXES` preserves the known correction or delta-review requirement and stops truthfully. `TEST_BLOCKED` preserves its blocker and records `WORKFLOW_BLOCKED`. Do not require a manifest/attestation package, reinterpret the testing result as a runtime crash, or recreate the skill's internal loops.
+Read the sibling `test-report.md` and `test-result.json`; they must agree. `TEST_PASS` proceeds to the fix report and retrospective. For `TEST_NEEDS_FIXES`, classify the concrete correction into an allowed `repair_scope`; safe in-budget repair re-enters Implement through the helper and requires fresh Review, configured verification, and Test, while unknown/unsafe scope or exhausted budget blocks. `TEST_BLOCKED` preserves its blocker and records `WORKFLOW_BLOCKED`. Do not require a manifest/attestation package, reinterpret the testing result as a runtime crash, or recreate the skill's internal loops.
 
 Only `TEST_PASS` may proceed to commit.
 
@@ -220,87 +231,25 @@ Only `TEST_PASS` may proceed to commit.
 
 ---
 
-### Phase 4: Commit & Learn
+### Phase 4: Report, Retrospective & Runtime Commit Handoff
 
-### Headless Commit Authorization
+**a. Fix report before commit preparation**
 
-This pipeline launch authorizes one local task commit only after Code Review and a consistent `TEST_PASS` report/result pair. Never ask a question, wait for input, or request confirmation. Invoke `/prizmkit-committer` with `mode=l4-headless`, `owner=prizmkit-l4`, and `local_commit_authorized=true`. The runtime controls any optional remote publication after the local commit; do not make a publication decision in this session.
+Write `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md` after `TEST_PASS`. Include the resolution summary, changed behavior, reproduction before/after, regression and review results, configured Browser/Manual evidence, durable knowledge captured, and acceptance-criteria checklist. Do not require a commit hash because the report must exist before the local commit.
 
-**Bug Fix Documentation Policy**:
-- **DEFAULT**: Run `/prizmkit-retrospective` with structural sync only (Job 1). Skip knowledge injection.
-- **Full retrospective** (Job 1 + Job 2): Only when the fix causes interface signature changes, dependency additions/removals, observable behavior changes, or reveals new TRAPs.
+Mark `bug-report` completed only after the report is valid.
 
-**a.** If a new pitfall was discovered (not previously in TRAPS):
-   - Update the affected module's TRAPS section in `.prizmkit/prizm-docs/`
-   - Format: `- [LOW|HIGH|CRITICAL] <description> | FIX: <approach>`
-   - Do not add dates.
+**b. Retrospective**
 
-**b.** Run `/prizmkit-retrospective` following the policy above.
-   Stage doc changes explicitly when `.prizmkit/prizm-docs/` changed.
+Invoke `/prizmkit-retrospective` with the exact artifact root, exact non-`.prizmkit/` `change_paths`, and a concise `change_summary`. Validate `outcome=RETRO_COMPLETE`; do not stage or commit.
 
-**c.** Stage all changed files explicitly (NEVER use `git add -A` or `git add .`):
-```bash
-git add <specific-files-modified>
-git add .prizmkit/prizm-docs/
-```
-
-**d.** Run `/prizmkit-committer`:
-   - Commit message prefix: `fix({{FIX_SCOPE}}): {{BUG_TITLE}}`
-   - Include both fix code and reproduction test
-   - The runtime controls any optional remote publication after the local commit; do not make a publication decision in this session
-
-**e.** Final verification:
-```bash
-git status --short
-```
-Working tree MUST be clean. If any files remain, amend the commit.
-
-**f.** Write fix report to `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md`:
+**c. Runtime commit preparation**
 
-The fix-report.md MUST contain:
-- **Bug Resolution Summary**: ID, title, status, phases completed
-- **What Was Fixed**: changes made, diff summary, commit hash
-- **Verification Results**: reproduction test before/after, regression tests, review findings
-
-{{IF_BROWSER_INTERACTION}}
-- **Browser Verification Results**: UI flows tested, browser tool used (if any), manual verification steps completed
-{{END_IF_BROWSER_INTERACTION}}
-
-- **Knowledge Captured**: TRAPS updated (if any), prevention recommendation
-- **Acceptance Criteria Verification**: checklist with pass/fail for each criterion
-
-- **Checkpoint update**: Set `prizmkit-retrospective`, `prizmkit-committer`, and `bug-report` to `completed` as each finishes.
+After this injected session validates every required artifact, invoke `/prizmkit-committer operation=prepare-runtime-commit artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/ evidence_paths=<exact paths> excluded_paths=.prizmkit/bugfix/{{BUG_ID}}/workflow-checkpoint.json request_path=.prizmkit/bugfix/{{BUG_ID}}/runtime-commit-request.json`. Require `COMMIT_REQUEST_READY`; this session records `COMMIT_PENDING` and stops. The Python runtime validates the request, executes/verifies the local commit, finalizes checkpoint `COMMITTED`, and controls any later publication.
 
 ---
 
-### Step 3: Report Session Status
-
-**CRITICAL**: Before exiting, write the session status file.
-
-Write to: `{{SESSION_STATUS_PATH}}`
-
-```json
-{
-  "session_id": "{{SESSION_ID}}",
-  "bug_id": "{{BUG_ID}}",
-  "status": "<success|partial|failed>",
-  "completed_phases": [1, 2, 3, 4],
-  "current_phase": 4,
-  "checkpoint_reached": "CP-4",
-  "fast_path": false,
-  "errors": [],
-  "can_resume": false,
-  "resume_from_phase": null,
-  "artifacts": {
-    "spec_path": ".prizmkit/bugfix/{{BUG_ID}}/spec.md",
-    "plan_path": ".prizmkit/bugfix/{{BUG_ID}}/plan.md",
-    "fix_report_path": ".prizmkit/bugfix/{{BUG_ID}}/fix-report.md"
-  },
-  "git_commit": "<commit hash>",
-  "traps_updated": false,
-  "timestamp": "{{TIMESTAMP}}"
-}
-```
+Runtime session status is written by the Python runtime. Do not write or edit `session-status.json`.
 
 ## Critical Paths
 
@@ -310,7 +259,6 @@ Write to: `{{SESSION_STATUS_PATH}}`
 | Spec | `.prizmkit/bugfix/{{BUG_ID}}/spec.md` |
 | Plan | `.prizmkit/bugfix/{{BUG_ID}}/plan.md` |
 | Fix Report | `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md` |
-| Session Status Output | {{SESSION_STATUS_PATH}} |
 | Project Root | {{PROJECT_ROOT}} |
 
 ## Reminders
@@ -321,5 +269,5 @@ Write to: `{{SESSION_STATUS_PATH}}`
 - plan.md first task MUST be reproduction test (RED → GREEN TDD)
 - Commit with `fix(<scope>):` prefix, NOT `feat:`
 - DEFAULT: `/prizmkit-retrospective` structural sync only (Job 1). Full retrospective when fix changes interfaces/dependencies/behavior
-- ALWAYS write session-status.json before exiting
-- Do NOT run `git add`/`git commit` during Phases 1-3 — all committed once in Phase 4
+- Runtime owns session-status.json; never hand-edit it
+- Do NOT run `git add`/`git commit`; the Python runtime executes the prepared local commit

package/bundled/dev-pipeline/templates/refactor-bootstrap-prompt.md

@@ -7,13 +7,12 @@
 - **Refactor Type**: {{REFACTOR_TYPE}}
 - **Priority**: {{PRIORITY}}
 - **Complexity**: {{COMPLEXITY}}
-- **Pipeline Mode**: {{PIPELINE_MODE}}
 
 ## Your Mission
 
 You are the **refactor session orchestrator**. Execute Refactor {{REFACTOR_ID}}: "{{REFACTOR_TITLE}}".
 
-**CRITICAL SESSION LIFECYCLE RULE**: You MUST NOT exit until all refactor work is complete, checkpoint state is updated, and session-status.json is written.
+**CRITICAL SESSION LIFECYCLE RULE**: Do not claim success until all refactor work and authoritative artifacts are complete. The Python runtime exclusively writes session status.
 
 **NON-INTERACTIVE MODE**: You are running in headless non-interactive mode. There is NO human on the other end. NEVER ask for user confirmation, NEVER wait for user input, and NEVER use interactive prompts. Resolve uncertainty from existing code, project docs, and conservative behavior-preserving defaults.
 
@@ -54,19 +53,28 @@ You are the **refactor session orchestrator**. Execute Refactor {{REFACTOR_ID}}:
 
 {{GLOBAL_CONTEXT}}
 
-## Refactor Artifacts Directory
+## Refactor Artifact Manifest
 
-Always use per-refactor subdirectory `.prizmkit/refactor/{{REFACTOR_ID}}/`:
+Always use `.prizmkit/refactor/{{REFACTOR_ID}}/`:
 
 ```
 .prizmkit/refactor/{{REFACTOR_ID}}/
 ├── spec.md
 ├── plan.md
+├── context-snapshot.md
 ├── review-report.md
+├── test-report.md
+├── test-result.json
 ├── refactor-report.md
-└── completion-summary.json
+├── completion-summary.json
+├── retrospective-result.json
+├── failure-log.md
+├── workflow-checkpoint.json
+└── runtime-commit-request.json
 ```
 
+Use `context-snapshot.md` for navigation, not as a substitute for current source, diffs, or authoritative artifacts. Verify writes with bounded reads/diffs/checks. `/prizmkit-test` exclusively owns the complete test repair loop. Do not stage, commit, amend, merge, push, reset, clean, or hand-edit runner-owned state.
+
 {{CHECKPOINT_SYSTEM}}
 
 ## Execution
@@ -85,7 +93,7 @@ Create refactor artifacts directory:
 
 ### Phase 1: Plan — Specification & Plan Generation
 
-Run `/prizmkit-plan` with `artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/`.
+Invoke `/prizmkit-plan` once with the exact artifact root, refactor requirement, and behavior-preservation context. It returns only `PLAN_READY` or `PLAN_BLOCKED`.
 
 The generated `spec.md` MUST include:
 - goals with acceptance criteria;
@@ -114,32 +122,28 @@ If `DEV_PORT` is unknown, discover it from project config; do not guess port 300
 
 {{END_IF_BROWSER_INTERACTION}}
 
-**Checkpoint update**: set step `prizmkit-plan` to `completed` in `{{CHECKPOINT_PATH}}`.
+**Checkpoint update**: this injected session maps `PLAN_READY` to a completed planning step after artifact validation and stops on `PLAN_BLOCKED`.
 
 ---
 
 ### Phase 3: Implement — Behavior-Preserving Refactoring
 
-Run `/prizmkit-implement` with `artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/`.
+Invoke `/prizmkit-implement` with the exact artifact root, reviewed spec/plan, and any explicit repair findings. It returns only `IMPLEMENTED` or `IMPLEMENT_BLOCKED`.
 
 Implementation requirements:
 1. Establish a green baseline before changes when tests exist.
 2. Execute plan tasks in order.
 3. Preserve existing behavior; do not mix unrelated feature work or bug fixes into the refactor.
-4. Run focused tests after each risky task and full relevant tests before review.
+4. Run the smallest focused checks needed during implementation; the authoritative affected-module test gate runs only after review and configured Browser verification.
 5. Mark each completed task `[x]` in plan.md.
 
-{{IF_BROWSER_INTERACTION}}
-Apply the browser behavior preservation protocol where UI behavior can be affected. Document evidence in implementation notes.
-{{END_IF_BROWSER_INTERACTION}}
-
-**Checkpoint update**: set step `prizmkit-implement` to `completed` in `{{CHECKPOINT_PATH}}`.
+**Checkpoint update**: this injected session maps `IMPLEMENTED` to completed after artifact/evidence validation and stops on `IMPLEMENT_BLOCKED`.
 
 ---
 
 ### Phase 4: Review — Code Review & Behavior Verification
 
-Run `/prizmkit-code-review` with `artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/` directly from the main orchestrator. The skill runs up to ten Main-Agent review rounds, directly repairs accepted findings, and verifies each repair. Missing required evidence, unsafe repair, failed verification, or round-ten non-convergence produces `NEEDS_FIXES`.
+Invoke `/prizmkit-code-review` with the exact artifact root, spec, plan, and complete current change. It returns only `PASS` or `NEEDS_FIXES`. The Skill runs up to ten Main-Agent review rounds, directly repairs accepted findings, and verifies each repair. Missing required evidence, unsafe repair, failed verification, or round-ten non-convergence produces `NEEDS_FIXES`.
 
 Review requirements:
 - verify refactor quality against spec.md and plan.md;
@@ -151,30 +155,16 @@ Review requirements:
 ```bash
 {{RUNTIME_HELPER_CMD}} text final-verdict .prizmkit/refactor/{{REFACTOR_ID}}/review-report.md
 ```
-Handle `REVIEW_PASS` and `REVIEW_NEEDS_FIXES` according to behavior-preservation policy. `REVIEW_INCOMPLETE` or `REVIEW_INVALID` means review did not complete: inspect progress once, write `failure-log.md`, and stop. Do not accept stale verdict text or replace required independent review with Main-Agent self-review.
+This injected session maps Skill `PASS`/`NEEDS_FIXES` to `REVIEW_PASS`/`REVIEW_NEEDS_FIXES`. `REVIEW_PASS` proceeds to configured verification. For `REVIEW_NEEDS_FIXES`, classify concrete findings into an allowed `repair_scope`; safe in-budget repair re-enters Implement through the helper, while unknown/unsafe scope or exhausted budget records `WORKFLOW_BLOCKED`. It must never proceed as a pass. `REVIEW_INCOMPLETE` or `REVIEW_INVALID` means review did not complete: inspect progress once, write `failure-log.md`, and stop. Do not accept stale verdict text or replace required independent review with Main-Agent self-review.
 
 **Checkpoint update**: set step `prizmkit-code-review` to `completed` only for an accepted valid Final Result. Do not complete it for incomplete or invalid reports.
 
 ---
 
-### Phase 5: Post-Review Test Gate — PrizmKit Test
-
-After a valid code-review PASS, run the mandatory test gate against the same artifact root:
-
-```text
-/prizmkit-test scope=this-change artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/
-```
-
-Behavior-preservation context remains supporting input. Read the sibling `test-report.md` and `test-result.json`; they must agree. `TEST_PASS` proceeds to browser verification or retrospective. `TEST_NEEDS_FIXES` preserves the known correction or delta-review requirement and stops truthfully. `TEST_BLOCKED` preserves its blocker and records `WORKFLOW_BLOCKED`. Do not require a manifest/attestation package, reinterpret the testing result as a runtime crash, or recreate the skill's internal loops.
-
-Only `TEST_PASS` may proceed to browser verification or commit.
-
----
-
 {{IF_BROWSER_INTERACTION}}
 ### Phase 5: Browser Verification
 
-Attempt browser verification for affected UI behavior. If no usable browser tool/server/port can be found and browser verification is non-blocking, mark browser verification `skipped` and continue. If explicit policy marks browser verification as blocking, mark it `failed` and write `failure-log.md`.
+After valid code-review PASS and before the authoritative Test stage, attempt browser verification for affected UI behavior. If no usable browser tool/server/port can be found and browser verification is non-blocking, mark browser verification `skipped` and continue. If explicit policy marks browser verification as blocking, mark it `failed`, write `failure-log.md`, and stop before Test.
 
 Record tool choice, URL, steps, screenshot path if available, result, and cleanup status in `.prizmkit/refactor/{{REFACTOR_ID}}/context-snapshot.md`.
 
@@ -183,25 +173,24 @@ Record tool choice, URL, steps, screenshot path if available, result, and cleanu
 ---
 
 {{END_IF_BROWSER_INTERACTION}}
-### Phase 6: Commit, Report & Completion Summary
+### Phase 5: Post-Review Test Gate — PrizmKit Test
 
-### Headless Commit Authorization
+After valid code-review PASS and configured Browser verification, run the mandatory test gate against the same artifact root:
 
-This pipeline launch authorizes one local task commit only after Code Review, a consistent `TEST_PASS` report/result pair, and required behavior-preservation gates pass. Never ask a question, wait for input, or request confirmation. Invoke `/prizmkit-committer` with `mode=l4-headless`, `owner=prizmkit-l4`, and `local_commit_authorized=true`. The runtime controls any optional remote publication after the local commit; do not make a publication decision in this session.
+```text
+/prizmkit-test scope=this-change artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/
+```
 
-Run `/prizmkit-retrospective` using the current working tree and staged changes as input context. Do not rely only on `git diff --cached`.
+Behavior-preservation context remains supporting input. Read the sibling `test-report.md` and `test-result.json`; they must agree. `TEST_PASS` proceeds to reports and retrospective. For `TEST_NEEDS_FIXES`, classify the concrete correction into an allowed `repair_scope`; safe in-budget repair re-enters Implement through the helper and requires fresh Review, configured verification, and Test, while unknown/unsafe scope or exhausted budget blocks. `TEST_BLOCKED` preserves its blocker and records `WORKFLOW_BLOCKED`. Do not require a manifest/attestation package, reinterpret the testing result as a runtime crash, or recreate the skill's internal loops.
 
-Stage changed files explicitly; never use `git add -A` or `git add .`.
+Only `TEST_PASS` may proceed to reports, retrospective, or commit preparation.
 
-Run `/prizmkit-committer` with commit prefix:
+---
 
-```text
-refactor({{REFACTOR_ID}}): {{REFACTOR_TITLE}}
-```
 
-The runtime controls any optional remote publication after this local commit; do not make a publication decision in this session.
+### Phase 6: Reports, Retrospective & Runtime Commit Handoff
 
-Write `.prizmkit/refactor/{{REFACTOR_ID}}/refactor-report.md` with:
+After `TEST_PASS`, write `.prizmkit/refactor/{{REFACTOR_ID}}/refactor-report.md` with:
 - refactor summary;
 - files/modules changed;
 - behavior-preservation evidence before/after;
@@ -219,36 +208,13 @@ Write `.prizmkit/refactor/{{REFACTOR_ID}}/completion-summary.json` for downstrea
 }
 ```
 
-**Checkpoint update**: set `prizmkit-retrospective`, `prizmkit-committer`, `refactor-report`, and `completion-summary` to `completed` as each finishes.
+Complete `refactor-report` and `completion-summary` before retrospective. Then invoke `/prizmkit-retrospective` with the exact artifact root, exact non-`.prizmkit/` `change_paths`, and a concise `change_summary`. Validate `outcome=RETRO_COMPLETE`; do not stage or commit.
 
----
-
-### Final: Report Session Status
+After this injected session validates every required artifact, invoke `/prizmkit-committer operation=prepare-runtime-commit artifact_dir=.prizmkit/refactor/{{REFACTOR_ID}}/ evidence_paths=<exact paths> excluded_paths=.prizmkit/refactor/{{REFACTOR_ID}}/workflow-checkpoint.json request_path=.prizmkit/refactor/{{REFACTOR_ID}}/runtime-commit-request.json`. Require `COMMIT_REQUEST_READY`; this session records `COMMIT_PENDING` and stops. The Python runtime executes/verifies the local commit, finalizes checkpoint `COMMITTED`, and controls any later publication.
 
-Before exiting, write `{{SESSION_STATUS_PATH}}`:
+---
 
-```json
-{
-  "session_id": "{{SESSION_ID}}",
-  "refactor_id": "{{REFACTOR_ID}}",
-  "status": "<success|partial|failed>",
-  "completed_phases": [1, 2, 3, 4, 5, 6],
-  "current_phase": 6,
-  "checkpoint_reached": "completion-summary",
-  "errors": [],
-  "can_resume": false,
-  "resume_from_phase": null,
-  "artifacts": {
-    "spec_path": ".prizmkit/refactor/{{REFACTOR_ID}}/spec.md",
-    "plan_path": ".prizmkit/refactor/{{REFACTOR_ID}}/plan.md",
-    "review_report_path": ".prizmkit/refactor/{{REFACTOR_ID}}/review-report.md",
-    "report_path": ".prizmkit/refactor/{{REFACTOR_ID}}/refactor-report.md",
-    "completion_summary_path": ".prizmkit/refactor/{{REFACTOR_ID}}/completion-summary.json"
-  },
-  "git_commit": "<commit hash>",
-  "behavior_preserved": true
-}
-```
+Runtime session status is written by the Python runtime. Do not write or edit `session-status.json`.
 
 ## Critical Paths
 
@@ -260,16 +226,15 @@ Before exiting, write `{{SESSION_STATUS_PATH}}`:
 | Review Report | `.prizmkit/refactor/{{REFACTOR_ID}}/review-report.md` |
 | Refactor Report | `.prizmkit/refactor/{{REFACTOR_ID}}/refactor-report.md` |
 | Completion Summary | `.prizmkit/refactor/{{REFACTOR_ID}}/completion-summary.json` |
-| Session Status Output | {{SESSION_STATUS_PATH}} |
 | Workflow Checkpoint | {{CHECKPOINT_PATH}} |
 | Project Root | {{PROJECT_ROOT}} |
 
 ## Reminders
 
-- Use the atomic skills: `/prizmkit-plan`, `/prizmkit-implement`, `/prizmkit-code-review`, `/prizmkit-retrospective`, `/prizmkit-committer`.
+- Use every formal atomic stage: `/prizmkit-plan`, `/prizmkit-implement`, `/prizmkit-code-review`, `/prizmkit-test`, `/prizmkit-retrospective`, `/prizmkit-committer` preparation.
 - Follow the active skills and runtime contracts; they determine execution topology, review ownership, and delegation boundaries
 - On execution-unit timeout, preserve the active checkout and artifacts and follow the owning contract's recovery or downgrade path
 - Behavior preservation is the first priority.
 - Browser verification is a mandatory attempt by default; it blocks only when explicitly required.
 - If `DEV_PORT` is unknown, discover it from project config; do not guess port 3000.
-- Always write session-status.json before exiting.
+- Runtime owns session-status.json; never hand-edit it.

package/bundled/dev-pipeline/templates/sections/artifact-manifest.md

@@ -0,0 +1,39 @@
+## Requirement Artifact Manifest
+
+Use one artifact root for the complete lifecycle: `{{ARTIFACT_DIR}}/`.
+
+```text
+{{ARTIFACT_DIR}}/
+├── spec.md
+├── plan.md
+├── context-snapshot.md
+├── review-report.md
+├── test-report.md
+├── test-result.json
+├── retrospective-result.json
+├── failure-log.md
+├── workflow-checkpoint.json
+├── runtime-commit-request.json
+{{FAMILY_ARTIFACTS}}
+```
+
+Artifact ownership:
+
+- `/prizmkit-plan` owns `spec.md` and `plan.md`.
+- The active implementation flow owns `context-snapshot.md` and implementation notes.
+- `/prizmkit-code-review` owns `review-report.md`.
+- `/prizmkit-test` owns the sibling `test-report.md` and `test-result.json` pair.
+- `/prizmkit-retrospective` owns `retrospective-result.json` and any justified `.prizmkit/prizm-docs/` updates.
+- The family orchestrator writes the family-specific reports/summaries listed above before retrospective and commit preparation.
+- `/prizmkit-committer` prepares `runtime-commit-request.json`; the Python runtime consumes it and executes the local commit.
+- `failure-log.md` records truthful blockers or incomplete exits.
+
+Runtime-state boundary:
+
+- Atomic Skills write only their stage-owned artifacts and domain results; they never read or mutate orchestration/checkpoint state.
+- This injected session orchestrator maps stage results and updates `workflow-checkpoint.json` through the declared helper.
+- Browser/process helpers may manage their declared PID files only through the runtime helper contract.
+- Never hand-edit family task state, session status, heartbeat, progress, retry, checkout, daemon, PID, or log files under `.prizmkit/state/`.
+- `session-status.json` is runtime-owned and is not part of this artifact manifest.
+
+Do not create duplicate lifecycle artifacts in another Feature, Bugfix, or Refactor directory.

package/bundled/dev-pipeline/templates/sections/bugfix-critical-paths.md

@@ -7,6 +7,8 @@
 | Plan | `.prizmkit/bugfix/{{BUG_ID}}/plan.md` |
 | Review Report | `.prizmkit/bugfix/{{BUG_ID}}/review-report.md` |
 | Fix Report | `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md` |
+| Test Result | `.prizmkit/bugfix/{{BUG_ID}}/test-result.json` |
+| Retrospective Result | `.prizmkit/bugfix/{{BUG_ID}}/retrospective-result.json` |
+| Runtime Commit Request | `.prizmkit/bugfix/{{BUG_ID}}/runtime-commit-request.json` |
 | Workflow Checkpoint | `{{CHECKPOINT_PATH}}` |
-| Session Status Output | `{{SESSION_STATUS_PATH}}` |
 | Project Root | `{{PROJECT_ROOT}}` |

package/bundled/dev-pipeline/templates/sections/bugfix-phase-commit-report.md

@@ -1,31 +1,14 @@
-### Phase 4: Commit & Learn
+### Bug Fix Report
 
-This phase follows successful automated, manual, or hybrid verification and runs only after a valid `TEST_PASS` report/result pair permits commit.
+After `TEST_PASS`, write `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md` before retrospective and runtime commit preparation.
 
-**Bug Fix Documentation Policy**:
-- Default: run `/prizmkit-retrospective` with structural sync only.
-- Full architecture knowledge capture only when the fix changes interfaces, dependencies, observable behavior, or reveals a durable TRAP.
-- TRAPS must use Prizm docs format: `- [LOW|HIGH|CRITICAL] <description> | FIX: <approach>`. Do not add dates.
+Include:
 
-Run `/prizmkit-retrospective` using the current working tree and staged changes as input context. Do not rely only on `git diff --cached`.
-
-Stage changed files explicitly; never use `git add -A` or `git add .`.
-
-Run `/prizmkit-committer` with commit prefix:
-
-```text
-fix({{FIX_SCOPE}}): {{BUG_TITLE}}
-```
-
-The runtime controls any optional remote publication after this local commit; do not make a publication decision in this session.
-
-Write `.prizmkit/bugfix/{{BUG_ID}}/fix-report.md` with:
 - root cause summary;
 - reproduction evidence;
-- fix summary;
-- files changed;
-- review/test/browser evidence;
-- manual/hybrid substitution evidence when applicable;
-- acceptance criteria checklist.
+- fix summary and files changed;
+- Code Review and Test results;
+- configured browser/manual/hybrid verification evidence and limitations;
+- final acceptance-criteria checklist.
 
-**Checkpoint update**: Set `prizmkit-retrospective`, `prizmkit-committer`, and `bug-report` to `completed` as each finishes.
+Do not stage or commit. Mark `bug-report` completed only after the report is current and complete.

package/bundled/dev-pipeline/templates/sections/bugfix-phase-diagnose-plan.md

@@ -14,10 +14,10 @@
    - Affected Files and Key Interfaces
    - Existing/Reproduction Tests
    - Relevant TRAPS/RULES
-4. Run `/prizmkit-plan` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`.
-5. Ensure `plan.md` starts with a reproduction test task that would fail before the fix.
+4. Invoke `/prizmkit-plan` once with the exact `artifact_dir`, bug requirement, diagnosis context, and project context. It returns only `PLAN_READY` or `PLAN_BLOCKED`.
+5. On `PLAN_READY`, validate that `plan.md` starts with a reproduction test task that would fail before the fix. On `PLAN_BLOCKED`, record the blocker and stop.
 6. Do not create a FAST_PATH that bypasses code review or the mandatory post-review test gate; all committable Bugfixes use the same atomic stages.
 
 **CP-BF-1**: `spec.md` and `plan.md` exist, and the first plan task is a reproduction test.
 
-**Checkpoint update**: Set step `bug-diagnosis-and-plan` to `completed`.
+**Checkpoint update**: This injected session sets `bug-diagnosis-and-plan` to `completed` with `stage_result=PLAN_READY` only after artifact validation. Record `PLAN_BLOCKED` as the failed planning result.

package/bundled/dev-pipeline/templates/sections/bugfix-phase-implement.md

@@ -2,7 +2,7 @@
 
 **Goal**: execute the fix plan with RED → GREEN evidence.
 
-Run `/prizmkit-implement` with `artifact_dir=.prizmkit/bugfix/{{BUG_ID}}/`:
+Invoke `/prizmkit-implement` with the exact `artifact_dir`, reviewed spec/plan, and any explicit repair findings. It returns only `IMPLEMENTED` or `IMPLEMENT_BLOCKED`:
 
 - Execute plan tasks in order.
 - First task creates or identifies a reproduction test that fails before the fix.
@@ -17,4 +17,4 @@ After implementation, verify:
 3. no in-scope regression remains unexplained;
 4. acceptance criteria have evidence.
 
-**Checkpoint update**: Set step `prizmkit-implement` to `completed`.
+**Checkpoint update**: This injected session maps `IMPLEMENTED` to a completed step only after the checks above. `IMPLEMENT_BLOCKED` records the blocker and stops.