aiwcli 0.10.3 → 0.11.1

package/dist/templates/cc-native/.claude/commands/cc-native/fresh-perspective.md

@@ -1,8 +0,0 @@
----
-description: Get an unbiased perspective on a problem without code context
----
-# Fresh Perspective
-
-Get a fresh, unbiased perspective on a problem by stripping away implementation details that could anchor your thinking to current patterns.
-
-Load and execute `_cc-native/workflows/fresh-perspective.md`.

package/dist/templates/cc-native/.windsurf/workflows/cc-native/fresh-perspective.md

@@ -1,8 +0,0 @@
----
-description: Get an unbiased perspective on a problem without code context
----
-# Fresh Perspective
-
-Get a fresh, unbiased perspective on a problem by stripping away implementation details that could anchor your thinking to current patterns.
-
-Load and execute `_cc-native/workflows/fresh-perspective.md`.

package/dist/templates/cc-native/MIGRATION.md

@@ -1,86 +0,0 @@
-# CC-Native v1.5.0 Migration Guide
-
-## Breaking Change: Context-Integrated Plan Review
-
-**Before (v1.4.0):** Reviews stored in `_output/cc-native/plans/{date}/{slug}/reviews/`
-**After (v1.5.0):** Reviews stored in `_output/contexts/{context_id}/reviews/`
-
-### What Changed
-
-1. **Removed `set_plan_state.py`** - No longer needed. Plan review now integrates with the shared context system.
-
-2. **Iteration state** - Now stored in `_output/contexts/{context_id}/reviews/iteration.json` instead of `~/.claude/plans/*.state.json`
-
-3. **Review output** - Written to context reviews folder when a context is active, with legacy fallback for backward compatibility.
-
-### Migration Steps
-
-**Automatic migration** - No action required. The hook will:
-- Use context reviews folder when an active context exists
-- Fall back to legacy path if no context found
-- Existing reviews remain in their original locations
-
----
-
-## v1.4.0 Migration (Previous)
-
-### Configuration File Rename
-
-**Before (v1.3.0):** `_cc-native/config.json`
-**After (v1.4.0):** `_cc-native/plan-review.config.json`
-
-```bash
-cd _cc-native
-mv config.json plan-review.config.json
-```
-
----
-
-## Features
-
-### Context-Integrated Reviews (v1.5.0)
-- Reviews written to `_output/contexts/{context_id}/reviews/`
-- Iteration state persists across sessions in context folder
-- Automatic context selection via session_id or single active context
-
-### Atomic Writes
-- Cross-platform atomic file writes (Windows + POSIX)
-- Retry logic with backoff
-- Error markers for debugging
-
-**No action required** - Enabled by default via `CC_NATIVE_ROBUST_WRITES=true`.
-
----
-
-## Upgrade Checklist
-
-- [ ] Test plan workflow: exit plan mode → review → archive
-- [ ] Verify reviews appear in `_output/contexts/{context_id}/reviews/`
-- [ ] Check iteration tracking persists across plan revisions
-
----
-
-## Rollback
-
-If you need to revert to v1.4.0:
-
-```bash
-git checkout v1.4.0 -- _cc-native/
-```
-
-**Note:** Existing reviews in context folders will remain but won't be used.
-
----
-
-## What Stays the Same
-
-- Workflow files (`_cc-native/workflows/*.md`)
-- Agent definitions (`.claude/agents/cc-native/`)
-- All slash commands (`/cc-native:*`)
-- Plan archiving behavior (`archive_plan.py`)
-
----
-
-## Questions?
-
-See TEMPLATE-SCHEMA.md for full documentation or report issues at the project repository.

package/dist/templates/cc-native/_cc-native/hooks/add_plan_context.py

@@ -1,130 +0,0 @@
-#!/usr/bin/env python3
-"""
-Plan context hook - handles question marking and advisory question nudges.
-
-Registered for two events:
-- PostToolUse: AskUserQuestion — marks that questions were asked this session.
-- PreToolUse: Task — nudges Plan subagent to ask questions first (advisory, never blocks).
-
-All enforcement is advisory: injects additionalContext to guide Claude toward asking
-non-obvious questions, but never blocks the tool call. Claude can proceed regardless.
-
-Fail-safe: Any error allows the action silently.
-"""
-
-import json
-import sys
-from pathlib import Path
-from typing import Any, Dict
-
-# Add lib directories to path for imports
-_hook_dir = Path(__file__).resolve().parent
-_cc_native_lib_dir = _hook_dir.parent / "lib"
-_shared_lib_dir = _hook_dir.parent.parent / "_shared" / "lib"
-sys.path.insert(0, str(_cc_native_lib_dir))
-sys.path.insert(0, str(_shared_lib_dir))
-
-from utils import (
-    mark_questions_asked,
-    was_questions_asked,
-)
-from base.hook_utils import emit_context
-from base.subprocess_utils import is_internal_call
-from base.logger import log_debug, log_info, log_warn, log_error, log_diagnostic
-from templates.plan_context import get_evaluation_context_reminder
-
-
-def is_plan_task(payload: Dict[str, Any]) -> bool:
-    """Check if this Task call spawns a Plan subagent."""
-    tool_input = payload.get("tool_input", {})
-    return str(tool_input.get("subagent_type", "")) == "Plan"
-
-
-CONTEXT_REMINDER = get_evaluation_context_reminder()
-
-
-def inject_evaluation_context() -> int:
-    """Inject evaluation context reminder without blocking."""
-    emit_context(CONTEXT_REMINDER)
-    return 0
-
-
-TASK_ENFORCEMENT_CONTEXT = (
-    "Before spawning a Plan agent, consider asking the user non-obvious questions "
-    "via AskUserQuestion. Code exploration reveals WHAT exists — questions reveal WHAT MATTERS.\n\n"
-    "Generate 5+ candidate questions across these categories, then keep only 3-4 where "
-    "different answers would lead to meaningfully different plans:\n\n"
-    "1. INTENT & SUCCESS CRITERIA — What does 'done well' look like? Are there multiple "
-    "interpretations of this request? What's a 10 vs a 6?\n\n"
-    "2. CONSTRAINTS & HISTORY — Has this been attempted before? Are there off-limits areas, "
-    "performance requirements, or security considerations not visible in the code?\n\n"
-    "3. TRADE-OFF PREFERENCES — Speed vs thoroughness? Minimal change vs clean architecture? "
-    "Backward compatibility vs clean break?\n\n"
-    "Frame each question with 2-3 concrete options so the user can choose rather than compose. "
-    "Use AskUserQuestion with structured options — never ask questions as inline text."
-)
-
-
-def nudge_task_questions() -> int:
-    """Advise Claude to ask questions before spawning Plan agent (never blocks)."""
-    emit_context(TASK_ENFORCEMENT_CONTEXT)
-    return 0
-
-
-def main() -> int:
-    # Guard: skip for internal subprocess calls (prevents recursive hook execution)
-    if is_internal_call():
-        return 0
-
-    try:
-        payload = json.load(sys.stdin)
-    except json.JSONDecodeError:
-        return 0  # Fail-safe
-
-    tool_name = payload.get("tool_name")
-    hook_event = payload.get("hook_event_name", "unknown")
-    log_diagnostic("add_plan_context", "receive", f"tool={tool_name}, event={hook_event}",
-                    inputs={"tool_name": tool_name, "hook_event": hook_event})
-
-    # PostToolUse: AskUserQuestion — mark that questions were asked
-    if tool_name == "AskUserQuestion":
-        session_id = str(payload.get("session_id", ""))
-        if session_id:
-            mark_questions_asked(session_id)
-            log_info("add_plan_context", f"Marked questions asked for session {session_id[:8]}...")
-        return 0
-
-    # PreToolUse: Task — nudge Plan subagent to ask questions first (advisory)
-    if tool_name == "Task":
-        if not is_plan_task(payload):
-            return 0  # Only gate Plan subagent spawns
-
-        permission_mode = payload.get("permission_mode", "")
-        if permission_mode != "plan":
-            return 0  # Only enforce during plan mode
-
-        session_id = payload.get("session_id")
-        if not session_id:
-            log_debug("add_plan_context", "No session_id for Task gate, skipping enforcement")
-            return 0
-
-        session_id_str = str(session_id)
-
-        if was_questions_asked(session_id_str):
-            log_info("add_plan_context", "Questions asked, allowing Plan Task with eval context")
-            log_diagnostic("add_plan_context", "decide", "Questions asked, allowing Plan Task",
-                            decision="allow_with_context", reasoning="was_questions_asked=True")
-            return inject_evaluation_context()
-
-        # Questions NOT asked: nudge toward asking questions (advisory only)
-        log_info("add_plan_context", "Questions not asked - nudging Plan Task to ask first")
-        log_diagnostic("add_plan_context", "decide", "Questions not asked, nudging Plan Task",
-                        decision="nudge", reasoning="was_questions_asked=False, advisory context")
-        return nudge_task_questions()
-
-    return 0
-
-
-if __name__ == "__main__":
-    from base.hook_utils import run_hook
-    run_hook(main, "add_plan_context")