prizmkit 1.1.152 → 1.1.154

package/bundled/skills/_metadata.json

@@ -1,5 +1,5 @@
 {
-  "version": "1.1.152",
+  "version": "1.1.154",
   "skills": {
     "prizmkit": {
       "description": "Framework introduction and navigation for the formal single-requirement lifecycle, project initialization, Prizm docs, and independent deployment.",

package/bundled/skills/bug-planner/references/schema-validation.md

@@ -17,7 +17,7 @@ Use this checklist for manual validation when `validate-bug-list.py` is not avai
 - [ ] `error_source.type`: one of `stack_trace`, `user_report`, `failed_test`, `log_pattern`, `monitoring_alert`
 - [ ] `verification_type`: one of `automated`, `manual`, `hybrid`
 - [ ] `acceptance_criteria`: non-empty array of strings
-- [ ] `status`: one of `pending`, `in_progress`, `completed`, `failed`, `skipped`, `needs_info`; new bugs should start as `pending`
+- [ ] `status`: one of `pending`, `in_progress`, `completed`, `failed`, `skipped`, `needs_info`, `auto_skipped`; new bugs should start as `pending`, while `auto_skipped` is runtime-managed for blocked descendants
 - [ ] `dependencies` if present: array of existing `B-NNN` IDs; no self-dependencies
 
 ## Consistency Checks

package/bundled/skills/bug-planner/scripts/validate-bug-list.py

@@ -29,7 +29,7 @@ from datetime import datetime, timezone
 VALID_SEVERITIES = {"critical", "high", "medium", "low"}
 VALID_SOURCE_TYPES = {"stack_trace", "user_report", "failed_test", "log_pattern", "monitoring_alert"}
 VALID_VERIFICATION_TYPES = {"automated", "manual", "hybrid"}
-VALID_STATUSES = {"pending", "in_progress", "completed", "failed", "skipped", "needs_info"}
+VALID_STATUSES = {"pending", "in_progress", "completed", "failed", "skipped", "needs_info", "auto_skipped"}
 BUG_ID_PATTERN = re.compile(r"^B-\d{3}$")
 SCHEMA_VERSION = "dev-pipeline-bug-fix-list-v1"
 

package/bundled/skills/bugfix-pipeline-launcher/SKILL.md

@@ -67,7 +67,7 @@ Only the start intent uses the execution-mode and configuration rounds.
 
 Use one `AskUserQuestion` call with exactly one question:
 
-- **Foreground (Recommended)** — build `bugfix run`; output remains attached to the user's terminal.
+- **Foreground (Recommended)** — build `run bugfix`; output remains attached to the user's terminal.
 - **Background daemon** — build `daemon bugfix start`; the runtime manages detached execution and logs.
 
 Wait for the response before asking configuration.
@@ -100,7 +100,7 @@ Foreground template:
 
 ```bash
 VERBOSE=1 MAX_RETRIES=3 \
-  python3 ./.prizmkit/dev-pipeline/cli.py bugfix run .prizmkit/plans/bug-fix-list.json
+  python3 ./.prizmkit/dev-pipeline/cli.py run bugfix .prizmkit/plans/bug-fix-list.json
 ```
 
 Background template:
@@ -118,7 +118,7 @@ Return:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py daemon bugfix status
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix status .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py status bugfix .prizmkit/plans/bug-fix-list.json
 ```
 
 Do not infer completion from the plan file alone when runtime state may exist.
@@ -151,39 +151,47 @@ Optionally include the daemon status command for the user to run afterward. Do n
 
 ## Retry Command
 
-Reset changes state only; it never starts execution and rejects `--run`. Always return reset and run as separate commands.
+Reset never starts execution and rejects `--run`. Every reset command must select exactly one mode: `--state-only`, `--fresh-checkout`, `--clean`, or `--preserve-runtime`. Always return reset and run as separate commands.
 
-A standard reset can discard uncommitted changes when the target task branch is currently checked out and records a fresh-checkout boundary. Clean reset additionally deletes the task branch and clears task history/artifacts. Preserve-runtime failed recovery instead scans the whole selected list, restores every failed root and its causal `auto_skipped` descendants, and retains valid checkout identity, sessions, checkpoints, artifacts, branches, worktrees, and WIP. If a recorded branch is missing, it creates a replacement from the recorded base, retains old artifacts for reference, and rewinds implementation-and-later checkpoint stages. Never hide these effects behind the word “retry.”
+State-only recovery resets selected failed/skipped/needs-info roots and only causal `auto_skipped` descendants whose complete blocker set is recovered. It resets retry/session/continuation state but never changes checkout identity, branches, worktrees, source changes, checkpoints, or artifact files. Selecting an auto-skipped bug includes all of its transitive blocking roots; `--all` recovers all eligible chains. Fresh-checkout performs the former standard reset, can discard uncommitted changes on the active task branch, and records a new checkout boundary. Clean additionally deletes the task branch and clears task history/artifacts. Preserve-runtime scans the whole list for failed roots, retains runtime history and valid checkout/WIP, and alone may replace a missing branch from the recorded base and rewind implementation-and-later checkpoints. Never hide these effects behind the word “retry.”
 
 Use `AskUserQuestion` before constructing retry commands:
 
-- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains in the selected list without cleanup, then run the list; disclose that this is whole-list recovery rather than a single-ID reset.
-- **Standard fresh retry** — reset one bug, then run that bug separately; warn that the active task branch can lose uncommitted changes.
+- **State-only causal recovery** — recover one bug closure, or all closures with `--all`, while preserving Git/checkouts/checkpoints/artifact files.
+- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains without cleanup while retaining history and continuation evidence; disclose that this is whole-list failed-only recovery.
+- **Fresh-checkout retry** — reset one bug, then run it separately; warn that the active task branch can lose uncommitted changes.
 - **Clean retry** — reset one bug, delete its task branch/history/artifacts, then run it separately.
 - **Cancel** — return no reset or run command.
 
+State-only template:
+
+```bash
+python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix B-001 --state-only .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix B-001 .prizmkit/plans/bug-fix-list.json
+```
+
 Preserve-runtime template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix --failed --preserve-runtime .prizmkit/plans/bug-fix-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix run .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix .prizmkit/plans/bug-fix-list.json
 ```
 
-Standard template:
+Fresh-checkout template:
 
 ```bash
-python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix B-001 .prizmkit/plans/bug-fix-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix run B-001 .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix B-001 --fresh-checkout .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix B-001 .prizmkit/plans/bug-fix-list.json
 ```
 
 Clean template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset bugfix B-001 --clean .prizmkit/plans/bug-fix-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix run B-001 .prizmkit/plans/bug-fix-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix B-001 .prizmkit/plans/bug-fix-list.json
 ```
 
-`--preserve-runtime` requires `--failed` and cannot be combined with `--clean`, an ID, a range, or another reset filter. Retry-budget flags such as `--max-retries` and `--max-infra-retries` belong on the run command, never reset. For clean retry, require explicit acknowledgement of branch deletion and possible uncommitted-change loss before displaying either command.
+State-only requires exactly one ID or explicit `--all` and rejects ranges/status filters. Fresh-checkout and clean accept one ID, range, or supported status filter. Preserve-runtime requires `--failed`, scans the whole list, and rejects IDs, ranges, `--all`, clean, other modes, and other filters. Retry-budget flags belong on the run command, never reset. For fresh-checkout or clean retry, require explicit acknowledgement of possible uncommitted-change loss; clean additionally requires acknowledgement of branch/history/artifact deletion.
 
 ## Output Contract
 

package/bundled/skills/feature-pipeline-launcher/SKILL.md

@@ -71,7 +71,7 @@ Only the start intent uses the execution-mode and configuration rounds.
 
 Use one `AskUserQuestion` call with exactly one question:
 
-- **Foreground (Recommended)** — build `feature run`; output remains attached to the user's terminal.
+- **Foreground (Recommended)** — build `run feature`; output remains attached to the user's terminal.
 - **Background daemon** — build `daemon feature start`; the runtime manages detached execution and logs.
 
 Wait for the response before asking configuration.
@@ -104,14 +104,14 @@ Foreground template:
 
 ```bash
 VERBOSE=1 MAX_RETRIES=3 \
-  python3 ./.prizmkit/dev-pipeline/cli.py feature run .prizmkit/plans/feature-list.json
+  python3 ./.prizmkit/dev-pipeline/cli.py run feature .prizmkit/plans/feature-list.json
 ```
 
 Foreground subset template:
 
 ```bash
 VERBOSE=1 MAX_RETRIES=3 \
-  python3 ./.prizmkit/dev-pipeline/cli.py feature run .prizmkit/plans/feature-list.json --features F-001:F-005
+  python3 ./.prizmkit/dev-pipeline/cli.py run feature .prizmkit/plans/feature-list.json --features F-001:F-005
 ```
 
 Background template:
@@ -136,7 +136,7 @@ Return both commands when the user asks for overall status:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py daemon feature status
-python3 ./.prizmkit/dev-pipeline/cli.py feature status .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py status feature .prizmkit/plans/feature-list.json
 ```
 
 Do not infer completion from the plan file alone when runtime state may exist.
@@ -169,39 +169,47 @@ Optionally include the daemon status command for the user to run afterward. Do n
 
 ## Retry Command
 
-Reset changes state only; it never starts execution and rejects `--run`. Always return reset and run as separate commands.
+Reset never starts execution and rejects `--run`. Every reset command must select exactly one mode: `--state-only`, `--fresh-checkout`, `--clean`, or `--preserve-runtime`. Always return reset and run as separate commands.
 
-A standard reset can discard uncommitted changes when the target task branch is currently checked out and records a fresh-checkout boundary. Clean reset additionally deletes the task branch and clears task history/artifacts. Preserve-runtime failed recovery instead scans the whole selected list, restores every failed root and its causal `auto_skipped` descendants, and retains valid checkout identity, sessions, checkpoints, artifacts, branches, worktrees, and WIP. If a recorded branch is missing, it creates a replacement from the recorded base, retains old artifacts for reference, and rewinds implementation-and-later checkpoint stages. Never hide these effects behind the word “retry.”
+State-only recovery resets selected failed/skipped roots and only causal `auto_skipped` descendants whose complete blocker set is recovered. It resets retry/session/continuation state but never changes checkout identity, branches, worktrees, source changes, checkpoints, or artifact files. Selecting an auto-skipped feature includes all of its transitive blocking roots; `--all` recovers all eligible chains. Fresh-checkout performs the former standard reset, can discard uncommitted changes on the active task branch, and records a new checkout boundary. Clean additionally deletes the task branch and clears task history/artifacts. Preserve-runtime scans the whole list for failed roots, retains runtime history and valid checkout/WIP, and alone may replace a missing branch from the recorded base and rewind implementation-and-later checkpoints. Never hide these effects behind the word “retry.”
 
 Use `AskUserQuestion` before constructing retry commands:
 
-- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains in the selected list without cleanup, then run the list; disclose that this is whole-list recovery rather than a single-ID reset.
-- **Standard fresh retry** — reset one feature, then run that feature separately; warn that the active task branch can lose uncommitted changes.
+- **State-only causal recovery** — recover one feature closure, or all closures with `--all`, while preserving Git/checkouts/checkpoints/artifact files.
+- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains without cleanup while retaining history and continuation evidence; disclose that this is whole-list failed-only recovery.
+- **Fresh-checkout retry** — reset one feature, then run it separately; warn that the active task branch can lose uncommitted changes.
 - **Clean retry** — reset one feature, delete its task branch/history/artifacts, then run it separately.
 - **Cancel** — return no reset or run command.
 
+State-only template:
+
+```bash
+python3 ./.prizmkit/dev-pipeline/cli.py reset feature F-001 --state-only .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run feature F-001 .prizmkit/plans/feature-list.json
+```
+
 Preserve-runtime template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset feature --failed --preserve-runtime .prizmkit/plans/feature-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py feature run .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run feature .prizmkit/plans/feature-list.json
 ```
 
-Standard template:
+Fresh-checkout template:
 
 ```bash
-python3 ./.prizmkit/dev-pipeline/cli.py reset feature F-001 .prizmkit/plans/feature-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py feature run F-001 .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py reset feature F-001 --fresh-checkout .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run feature F-001 .prizmkit/plans/feature-list.json
 ```
 
 Clean template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset feature F-001 --clean .prizmkit/plans/feature-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py feature run F-001 .prizmkit/plans/feature-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run feature F-001 .prizmkit/plans/feature-list.json
 ```
 
-`--preserve-runtime` requires `--failed` and cannot be combined with `--clean`, an ID, a range, or another reset filter. Retry-budget flags such as `--max-retries` and `--max-infra-retries` belong on the run command, never reset. For clean retry, require explicit acknowledgement of branch deletion and possible uncommitted-change loss before displaying either command.
+State-only requires exactly one ID or explicit `--all` and rejects ranges/status filters. Fresh-checkout and clean accept one ID, range, or supported status filter. Preserve-runtime requires `--failed`, scans the whole list, and rejects IDs, ranges, `--all`, clean, other modes, and other filters. Retry-budget flags belong on the run command, never reset. For fresh-checkout or clean retry, require explicit acknowledgement of possible uncommitted-change loss; clean additionally requires acknowledgement of branch/history/artifact deletion.
 
 ## Output Contract
 

package/bundled/skills/recovery-workflow/SKILL.md

@@ -7,12 +7,14 @@ description: "Return the exact PrizmKit CLI command that continues an interrupte
 
 Find an interrupted pipeline task's durable checkout record and return the existing runtime command that continues that task on its recorded branch. The runtime already owns branch reuse, checkpoint interpretation, session continuation, and task execution; this skill must not reproduce those mechanisms.
 
+This skill invokes its own bundled detector. The canonical runtime exposes no separate Recovery command namespace, prompt generator, or Recovery AI-session launcher.
+
 ## Boundary
 
 This workflow:
 - Detects active feature, bugfix, and refactor task checkout records.
 - Confirms that each recorded task branch still exists.
-- Returns the exact item-specific `cli.py ... run <TASK_ID>` command.
+- Returns the exact item-specific `cli.py run <family> <TASK_ID>` command.
 - Shows the branch, task family, task ID, and checkout record that justify the command.
 
 This workflow does not:
@@ -71,9 +73,9 @@ Report that no active interrupted pipeline task branch was found. Do not fall ba
 Return the detector-provided command verbatim. The canonical forms are:
 
 ```bash
-python3 ./.prizmkit/dev-pipeline/cli.py feature run F-001
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix run B-001
-python3 ./.prizmkit/dev-pipeline/cli.py refactor run R-001
+python3 ./.prizmkit/dev-pipeline/cli.py run feature F-001
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix B-001
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor R-001
 ```
 
 Do not prepend `git checkout`. The runtime resolves the durable checkout identity and continues the task on its recorded branch or linked worktree.
@@ -91,7 +93,7 @@ task_id: <F-NNN | B-NNN | R-NNN>
 branch: <recorded active_dev_branch>
 checkout: <project-relative checkout.json path>
 command:
-python3 ./.prizmkit/dev-pipeline/cli.py <family> run <TASK_ID>
+python3 ./.prizmkit/dev-pipeline/cli.py run <family> <TASK_ID>
 
 execution: not performed
 ```

package/bundled/skills/recovery-workflow/evals/evals.json

@@ -6,7 +6,7 @@
       "id": 1,
       "name": "single-feature-interruption",
       "prompt": "My feature pipeline process stopped. Give me the command to continue it.",
-      "expected_output": "Detect one active feature checkout record with an existing branch and return python3 ./.prizmkit/dev-pipeline/cli.py feature run F-001 without executing it.",
+      "expected_output": "Detect one active feature checkout record with an existing branch and return python3 ./.prizmkit/dev-pipeline/cli.py run feature F-001 without executing it.",
       "setup_description": "Create .prizmkit/state/features/F-001/checkout.json with state active and an existing dev/F-001-* local branch.",
       "files": []
     },
@@ -14,7 +14,7 @@
       "id": 2,
       "name": "single-bugfix-interruption",
       "prompt": "Resume the interrupted bugfix task.",
-      "expected_output": "Detect the active bugfix checkout and return python3 ./.prizmkit/dev-pipeline/cli.py bugfix run B-001 without reconstructing phases or executing the command.",
+      "expected_output": "Detect the active bugfix checkout and return python3 ./.prizmkit/dev-pipeline/cli.py run bugfix B-001 without reconstructing phases or executing the command.",
       "setup_description": "Create .prizmkit/state/bugfix/B-001/checkout.json with state active and an existing bugfix/B-001-* local branch.",
       "files": []
     },

package/bundled/skills/recovery-workflow/references/detection.md

@@ -24,9 +24,9 @@ Ignore records whose state is `completed` or `reset`. Also ignore session logs,
 The item-specific foreground command reloads the active checkout record and reuses its branch or linked worktree:
 
 ```bash
-python3 ./.prizmkit/dev-pipeline/cli.py feature run <F-NNN>
-python3 ./.prizmkit/dev-pipeline/cli.py bugfix run <B-NNN>
-python3 ./.prizmkit/dev-pipeline/cli.py refactor run <R-NNN>
+python3 ./.prizmkit/dev-pipeline/cli.py run feature <F-NNN>
+python3 ./.prizmkit/dev-pipeline/cli.py run bugfix <B-NNN>
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor <R-NNN>
 ```
 
 The command does not need an explicit branch argument. Do not prepend `git checkout`, because the runtime owns checkout identity and branch/worktree reuse.

package/bundled/skills/recovery-workflow/scripts/detect-recovery-state.py

@@ -66,7 +66,7 @@ def current_branch(project_root: Path) -> str:
 
 
 def command_for(family: dict[str, str], task_id: str) -> str:
-    return f"python3 ./.prizmkit/dev-pipeline/cli.py {family['name']} run {task_id}"
+    return f"python3 ./.prizmkit/dev-pipeline/cli.py run {family['name']} {task_id}"
 
 
 def detect_candidates(project_root: Path) -> list[dict[str, object]]:

package/bundled/skills/refactor-pipeline-launcher/SKILL.md

@@ -79,7 +79,7 @@ If the project instruction files or package metadata name a baseline test comman
 
 Use one `AskUserQuestion` call with exactly one question:
 
-- **Foreground (Recommended)** — build `refactor run`; output remains attached to the user's terminal.
+- **Foreground (Recommended)** — build `run refactor`; output remains attached to the user's terminal.
 - **Background daemon** — build `daemon refactor start`; the runtime manages detached execution and logs.
 
 Wait for the response before asking configuration.
@@ -113,7 +113,7 @@ Foreground template:
 
 ```bash
 VERBOSE=1 MAX_RETRIES=3 STRICT_BEHAVIOR_CHECK=1 \
-  python3 ./.prizmkit/dev-pipeline/cli.py refactor run .prizmkit/plans/refactor-list.json
+  python3 ./.prizmkit/dev-pipeline/cli.py run refactor .prizmkit/plans/refactor-list.json
 ```
 
 Background template:
@@ -131,7 +131,7 @@ Return:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py daemon refactor status
-python3 ./.prizmkit/dev-pipeline/cli.py refactor status .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py status refactor .prizmkit/plans/refactor-list.json
 ```
 
 Do not infer completion from the plan file alone when runtime state may exist.
@@ -164,39 +164,47 @@ Optionally include the daemon status command for the user to run afterward. Do n
 
 ## Retry Command
 
-Reset changes state only; it never starts execution and rejects `--run`. Always return reset and run as separate commands.
+Reset never starts execution and rejects `--run`. Every reset command must select exactly one mode: `--state-only`, `--fresh-checkout`, `--clean`, or `--preserve-runtime`. Always return reset and run as separate commands.
 
-A standard reset can discard uncommitted changes when the target task branch is currently checked out and records a fresh-checkout boundary. Clean reset additionally deletes the task branch and clears task history/artifacts. Preserve-runtime failed recovery instead scans the whole selected list, restores every failed root and its causal `auto_skipped` descendants, and retains valid checkout identity, sessions, checkpoints, artifacts, branches, worktrees, and WIP. If a recorded branch is missing, it creates a replacement from the recorded base, retains old artifacts for reference, and rewinds implementation-and-later checkpoint stages. Never hide these effects behind the word “retry.”
+State-only recovery resets selected failed/skipped roots and only causal `auto_skipped` descendants whose complete blocker set is recovered. It resets retry/session/continuation state but never changes checkout identity, branches, worktrees, source changes, checkpoints, or artifact files. Selecting an auto-skipped refactor includes all of its transitive blocking roots; `--all` recovers all eligible chains. Fresh-checkout performs the former standard reset, can discard uncommitted changes on the active task branch, and records a new checkout boundary. Clean additionally deletes the task branch and clears task history/artifacts. Preserve-runtime scans the whole list for failed roots, retains runtime history and valid checkout/WIP, and alone may replace a missing branch from the recorded base and rewind implementation-and-later checkpoints. Never hide these effects behind the word “retry.”
 
 Use `AskUserQuestion` before constructing retry commands:
 
-- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains in the selected list without cleanup, then run the list; disclose that this is whole-list recovery rather than a single-ID reset.
-- **Standard fresh retry** — reset one refactor, then run that refactor separately; warn that the active task branch can lose uncommitted changes.
+- **State-only causal recovery** — recover one refactor closure, or all closures with `--all`, while preserving Git/checkouts/checkpoints/artifact files.
+- **Preserve failed runtime (Recommended for continuing failed work)** — recover all failed chains without cleanup while retaining history and continuation evidence; disclose that this is whole-list failed-only recovery.
+- **Fresh-checkout retry** — reset one refactor, then run it separately; warn that the active task branch can lose uncommitted changes.
 - **Clean retry** — reset one refactor, delete its task branch/history/artifacts, then run it separately.
 - **Cancel** — return no reset or run command.
 
+State-only template:
+
+```bash
+python3 ./.prizmkit/dev-pipeline/cli.py reset refactor R-001 --state-only .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor R-001 .prizmkit/plans/refactor-list.json
+```
+
 Preserve-runtime template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset refactor --failed --preserve-runtime .prizmkit/plans/refactor-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py refactor run .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor .prizmkit/plans/refactor-list.json
 ```
 
-Standard template:
+Fresh-checkout template:
 
 ```bash
-python3 ./.prizmkit/dev-pipeline/cli.py reset refactor R-001 .prizmkit/plans/refactor-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py refactor run R-001 .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py reset refactor R-001 --fresh-checkout .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor R-001 .prizmkit/plans/refactor-list.json
 ```
 
 Clean template:
 
 ```bash
 python3 ./.prizmkit/dev-pipeline/cli.py reset refactor R-001 --clean .prizmkit/plans/refactor-list.json
-python3 ./.prizmkit/dev-pipeline/cli.py refactor run R-001 .prizmkit/plans/refactor-list.json
+python3 ./.prizmkit/dev-pipeline/cli.py run refactor R-001 .prizmkit/plans/refactor-list.json
 ```
 
-`--preserve-runtime` requires `--failed` and cannot be combined with `--clean`, an ID, a range, or another reset filter. Retry-budget flags such as `--max-retries` and `--max-infra-retries` belong on the run command, never reset. For clean retry, require explicit acknowledgement of branch deletion and possible uncommitted-change loss before displaying either command.
+State-only requires exactly one ID or explicit `--all` and rejects ranges/status filters. Fresh-checkout and clean accept one ID, range, or supported status filter. Preserve-runtime requires `--failed`, scans the whole list, and rejects IDs, ranges, `--all`, clean, other modes, and other filters. Retry-budget flags belong on the run command, never reset. For fresh-checkout or clean retry, require explicit acknowledgement of possible uncommitted-change loss; clean additionally requires acknowledgement of branch/history/artifact deletion.
 
 ## Output Contract
 

package/bundled/templates/project-memory-template.md

@@ -44,18 +44,22 @@ Placeholders: `<family> = feature | bugfix | refactor`; `<task-id> = F-NNN | B-N
 
 ```bash
 # Run/resume the whole list or one task
-python3 ./.prizmkit/dev-pipeline/cli.py <family> run [<list>] [--max-retries N] [--max-infra-retries N]
-python3 ./.prizmkit/dev-pipeline/cli.py <family> run <task-id> [<list>] [--max-retries N] [--max-infra-retries N]
+python3 ./.prizmkit/dev-pipeline/cli.py run <family> [<list>] [--max-retries N] [--max-infra-retries N]
+python3 ./.prizmkit/dev-pipeline/cli.py run <family> <task-id> [<list>] [--max-retries N] [--max-infra-retries N]
 
-# Inspect or recover task state
-python3 ./.prizmkit/dev-pipeline/cli.py <family> status [<list>]
-python3 ./.prizmkit/dev-pipeline/cli.py <family> unskip [<task-id>] [<list>]
-python3 ./.prizmkit/dev-pipeline/cli.py recovery detect
+# Inspect task state
+python3 ./.prizmkit/dev-pipeline/cli.py status <family> [<list>]
+python3 ./.prizmkit/dev-pipeline/cli.py status all
 
-# Reset state (never starts execution)
-python3 ./.prizmkit/dev-pipeline/cli.py reset <family> <task-id> [<list>]
+# Inspect runtime and AI CLI resolution
+python3 ./.prizmkit/dev-pipeline/cli.py diagnostics ai-cli
+
+# Reset state (exactly one explicit mode; never starts execution)
+python3 ./.prizmkit/dev-pipeline/cli.py reset <family> <task-id> --state-only [<list>]
+python3 ./.prizmkit/dev-pipeline/cli.py reset <family> --state-only --all [<list>]
+python3 ./.prizmkit/dev-pipeline/cli.py reset <family> <task-id-or-range> --fresh-checkout [<list>]
+python3 ./.prizmkit/dev-pipeline/cli.py reset <family> <task-id-or-range> --clean [<list>]
 python3 ./.prizmkit/dev-pipeline/cli.py reset <family> --failed --preserve-runtime [<list>]
-python3 ./.prizmkit/dev-pipeline/cli.py reset <family> <task-id> --clean [<list>]
 
 # Background operation
 python3 ./.prizmkit/dev-pipeline/cli.py daemon <family> start [<list>]
@@ -64,10 +68,14 @@ python3 ./.prizmkit/dev-pipeline/cli.py daemon <family> logs [--lines N|--follow
 python3 ./.prizmkit/dev-pipeline/cli.py daemon <family> stop
 ```
 
+Interrupted-task recovery: invoke `recovery-workflow`. Its read-only detector reports active checkout candidates and returns the ordinary `python3 ./.prizmkit/dev-pipeline/cli.py run <family> <task-id>` command without executing it.
+
 Runtime boundaries:
 - Reset and run are separate operations; reset rejects `--run`. Retry ceilings belong to `run`, not reset.
-- `--preserve-runtime` requires `--failed`, scans the whole selected list, and cannot be combined with `--clean`, an ID/range, or another reset filter.
-- Preserve-runtime recovery keeps valid checkout identity, sessions, checkpoints, artifacts, branches, worktrees, and WIP. If the recorded branch is missing, it creates a replacement from the recorded base and rewinds implementation-and-later checkpoint stages while retaining old artifacts for reference. Clean reset is destructive.
+- Exactly one of `--state-only`, `--fresh-checkout`, `--clean`, or `--preserve-runtime` is required. Missing/conflicting modes fail before state or Git mutation.
+- State-only accepts one ID or explicit `--all`; it resets causal dependency state while preserving checkout identity, branches/worktrees, source changes, checkpoints, and artifact files.
+- Fresh-checkout and clean accept an ID, range, or supported status filter. Fresh-checkout publishes a new checkout boundary; clean additionally deletes branch/history/artifact resources.
+- Preserve-runtime requires `--failed`, scans the whole list, and rejects IDs, ranges, `--all`, clean, other modes, and other filters. It retains runtime history and valid checkout/WIP; a missing branch is replaced from the recorded base with implementation-and-later checkpoint rewind.
 - In zsh/Bash multiline commands, `\` must be the final character on the line with no trailing space.
 
 ### Lightweight Planning Path

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "prizmkit",
-  "version": "1.1.152",
+  "version": "1.1.154",
   "description": "Create a new PrizmKit-powered project with clean initialization — no framework dev files, just what you need.",
   "type": "module",
   "bin": {