pactkit 2.14.1__tar.gz → 2.14.2__tar.gz

pactkit-2.14.2/.opencode/pactkit.yaml

@@ -0,0 +1,23 @@
+# PactKit Configuration (OpenCode)
+# Only override non-default values. Absent keys = default (all enabled).
+
+stack: python
+developer: "slim"
+
+ci:
+  provider: github
+
+issue_tracker:
+  provider: github
+
+release:
+  github_release: true
+
+visualize:
+  graph_provider: codegraph
+
+agent_models:
+  code-explorer: haiku
+  product-designer: opus
+  system-architect: opus
+  system-medic: haiku

{pactkit-2.14.1 → pactkit-2.14.2}/CHANGELOG.md

@@ -4,6 +4,20 @@ All notable changes to PactKit will be documented in this file.
 
 Format follows [Keep a Changelog](https://keepachangelog.com/).
 
+## [2.14.2] - 2026-06-01
+
+### Added
+- **Managed-block update for project CLAUDE.md** (STORY-slim-127) — `pactkit update` now uses `<!-- pactkit:start -->` / `<!-- pactkit:end -->` markers. User content outside the managed block is preserved across updates. Supports four migration paths: fresh install, existing markers, legacy PactKit template, and user-modified files.
+- **Codegraph sync enforcement via code** (STORY-slim-126) — `pactkit visualize --lazy` and `pactkit sync` now run `codegraph sync` automatically when `.codegraph/` exists. Removed prompt-based instructions that relied on AI compliance.
+- **Codegraph priority in generated CLAUDE.md** — When `.codegraph/` exists, the generated project CLAUDE.md includes a "Code Intelligence" section instructing AI to prefer codegraph over grep/find.
+
+### Changed
+- **MCP strategy trimmed to Context7 + Memory** — Removed Playwright, Chrome DevTools, Draw.io, and shadcn from MCP rules and recommendations. These remain as conditional no-ops in command prompts but are no longer actively promoted.
+- **Config backfill removed** — `pactkit update` no longer backfills default sections into `pactkit.yaml`. Absent keys now mean "accept default", keeping config files minimal.
+
+### Fixed
+- **Stale rule warnings** — Removed obsolete rule names from `.opencode/pactkit.yaml` that caused "Unknown rule" warnings on every `pactkit update`.
+
 ## [2.14.1] - 2026-05-28
 
 ### Fixed

{pactkit-2.14.1 → pactkit-2.14.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pactkit
-Version: 2.14.1
+Version: 2.14.2
 Summary: Spec-driven agentic DevOps toolkit for AI coding assistants
 Project-URL: Homepage, https://pactkit.dev
 Project-URL: Repository, https://github.com/pactkit/pactkit-public
@@ -123,7 +123,7 @@ Requires Python 3.10+ and one of:
 # Deploy to all 3 IDEs at once
 pactkit init
 
-# Update to latest playbooks (preserves your config)
+# Update to latest playbooks (preserves your custom content in CLAUDE.md)
 pactkit update
 ```
 

{pactkit-2.14.1 → pactkit-2.14.2}/README.md

@@ -81,7 +81,7 @@ Requires Python 3.10+ and one of:
 # Deploy to all 3 IDEs at once
 pactkit init
 
-# Update to latest playbooks (preserves your config)
+# Update to latest playbooks (preserves your custom content in CLAUDE.md)
 pactkit update
 ```
 

{pactkit-2.14.1 → pactkit-2.14.2}/docs/architecture/governance/archive/lessons_archive_202603.md

@@ -62,3 +62,4 @@
 | 2026-03-26 | _atomic_mmd_write() in visualize.py prevents truncated .mmd on crash via tmp+rename; all 4 write sites converted | visualize.py:_atomic_mmd_write |
 | 2026-03-26 | test_story_slim056.py _init_project() helper creates realistic pactkit project fixture for subprocess E2E tests; 60 tests cover all 25 subcommands | test_story_slim056.py:_init_project |
 | 2026-03-26 | entry_points auto-discovery with _load_entry_point_deployers() enables pip install pactkit-opencode to self-register; pre-existing tests referencing extracted functions must be updated to import from adapter package | deployer.py:_load_entry_point_deployers |
+| 2026-03-26 | VALID_FORMATS auto-derives from FORMAT_PROFILES.keys() — removing a profile cascades to config, CLI, and deployer with zero manual sync | profiles.py:FORMAT_PROFILES |

{pactkit-2.14.1 → pactkit-2.14.2}/docs/architecture/governance/lessons.md

@@ -2,7 +2,6 @@
 
 | Date | Lesson | Context |
 |------|--------|---------|
-| 2026-03-26 | VALID_FORMATS auto-derives from FORMAT_PROFILES.keys() — removing a profile cascades to config, CLI, and deployer with zero manual sync | profiles.py:FORMAT_PROFILES |
 | 2026-03-26 | DeployerBase static methods use lazy imports to avoid circular dependencies between deploy_base.py and deployer.py | deploy_base.py:DeployerBase.render_prompt |
 | 2026-03-26 | Thin adapter conversion: pactkit-codex reduced from 15,356 lines (full fork) to 668 lines (thin adapter) by importing core via DeployerBase. Key: deploy_codex_playbooks needs explicit path replacement for ALL format prefixes (~/.claude/, ~/.config/opencode/ → ~/.codex/) | pactkit_codex/deployer.py:deploy_codex_playbooks |
 | 2026-03-27 | When migrating commands to skills (subdir/SKILL.md format), all path assertions in pre-existing tests must be updated in the same commit — 17 tests broke because they asserted flat commands/*.md paths | deployer.py:_deploy_commands |
@@ -52,3 +51,5 @@
 | 2026-05-25 | When SCAN_EXCLUDES is a module-level constant shared across graph modes, add mode-specific exclusion logic at the call site rather than modifying the constant — pass call_extra_excludes = SCAN_EXCLUDES - {'tests'} only for call mode to avoid affecting file/class modes | visualize.py:_build_call_graph |
 | 2026-05-25 | _write_sqlite_db reuses func_registry and rel_edges already in memory from _build_call_graph — no second AST scan; pactkit query reads db directly without touching pactkit.yaml | visualize.py:_build_call_graph |
 | 2026-05-26 | SKILL.md model: frontmatter is passed through by deployer without transformation — source prompts in commands.py/workflows.py are the single source of truth for deployed skill metadata | commands.py:COMMANDS_CONTENT |
+| 2026-06-01 | codegraph sync must be enforced by code (pactkit visualize/sync), not by prompt instructions — prompt-only enforcement is unreliable for deterministic operations | lazy_visualize.py:codegraph_sync |
+| 2026-06-01 | Managed-block pattern (start/end markers + regex replace) is the canonical way to update mixed-ownership files — same pattern for CLAUDE.md and CLAUDE.local.md | deployer.py:_upsert_claude_md_managed_block |

{pactkit-2.14.1 → pactkit-2.14.2}/docs/architecture/governance/rules.md

@@ -15,7 +15,7 @@
 
 ## Invariants
 
-1. All 3986+ tests must pass before any commit to `main`.
+1. All 4009+ tests must pass before any commit to `main`.
 2. Specs (`docs/specs/`) are the source of truth — code conforms to specs, not the reverse.
 3. CLI entry point is `pactkit` via `src/pactkit/cli.py:main`.
 4. No runtime dependencies beyond `pyyaml`.

pactkit-2.14.2/docs/product/archive/archive_202606.md

@@ -0,0 +1,17 @@
+
+### [STORY-slim-126] codegraph sync code enforcement
+> Spec: docs/specs/STORY-slim-126.md
+
+- [x] Add codegraph_sync() function
+- [x] Integrate into visualize path
+- [x] Add pactkit sync subcommand
+- [x] Remove prompt instructions
+- [x] Unit tests
+
+### [STORY-slim-127] project CLAUDE.md managed-block update
+> Spec: docs/specs/STORY-slim-127.md
+
+- [x] Add marker constants
+- [x] Extract managed content builder
+- [x] Rewrite _generate_project_claude_md with upsert
+- [x] Unit tests

{pactkit-2.14.1 → pactkit-2.14.2}/docs/product/context.md

@@ -1,5 +1,5 @@
 # Project Context (Auto-generated)
-> Last updated: 2026-05-28T18:15:01+08:00 by pactkit context
+> Last updated: 2026-06-01T11:05:34+08:00 by pactkit context
 
 ## Sprint Status
 Backlog: 3 | In Progress: 0 | Done: 0 stories
@@ -16,11 +16,11 @@ None
 * main
 
 ## Key Decisions
-- Lifecycle gap pattern: when artifact has create+consume but no update mechanism, add conditional sync in the modifying command. Applied: Act Phase 4 Journey Sync step in commands.py COMMANDS_CONTENT['project-act.md']
-- code_graph.mmd uses sanitized node IDs (src_pactkit_generators_deployer_py), not bare filenames — grep patterns for fan-in/fan-out must use .* wildcard or they silently never match
 - When SCAN_EXCLUDES is a module-level constant shared across graph modes, add mode-specific exclusion logic at the call site rather than modifying the constant — pass call_extra_excludes = SCAN_EXCLUDES - {'tests'} only for call mode to avoid affecting file/class modes
 - _write_sqlite_db reuses func_registry and rel_edges already in memory from _build_call_graph — no second AST scan; pactkit query reads db directly without touching pactkit.yaml
 - SKILL.md model: frontmatter is passed through by deployer without transformation — source prompts in commands.py/workflows.py are the single source of truth for deployed skill metadata
+- codegraph sync must be enforced by code (pactkit visualize/sync), not by prompt instructions — prompt-only enforcement is unreliable for deterministic operations
+- Managed-block pattern (start/end markers + regex replace) is the canonical way to update mixed-ownership files — same pattern for CLAUDE.md and CLAUDE.local.md
 
 ## Next Recommended Action
 `/project-plan`

pactkit-2.14.2/docs/specs/STORY-slim-126.md

@@ -0,0 +1,131 @@
+# STORY-slim-126: codegraph sync code enforcement
+
+| Field | Value |
+|-------|-------|
+| ID | STORY-slim-126 |
+| Status | Done |
+| Priority | P1 |
+| Release | 2.15.0 |
+
+## Background
+
+`codegraph sync` is currently enforced only via prompt instructions in 4 PDCA skill templates (Act Phase 4, Done Phase 2, Hotfix Phase 3.6, Plan Phase 1 auto-setup). This violates the "Code Enforces, Prompt Instructs" principle — if the AI ignores the prompt instruction, the codegraph index becomes stale and `pactkit query` returns outdated results.
+
+The fix: integrate `codegraph sync` into the `pactkit visualize --lazy` code path so it runs automatically whenever source files changed. Remove the corresponding prompt instructions afterward.
+
+## Requirements
+
+### R1: Auto-sync codegraph in visualize path (MUST)
+
+When `pactkit visualize --lazy` determines that source files changed (should_visualize returns True) AND `.codegraph/` directory exists in project root AND `codegraph` is on `$PATH`, automatically run `codegraph sync` after graph generation completes.
+
+### R2: Standalone sync function (MUST)
+
+Extract a reusable `codegraph_sync(project_root)` function in `lazy_visualize.py` that:
+- Checks `.codegraph/` exists
+- Checks `codegraph` binary is available via `shutil.which`
+- Runs `codegraph sync <path>` with subprocess
+- Returns a status tuple: `(synced: bool, message: str)`
+- Fails silently (returns False + reason) when codegraph is unavailable
+
+### R3: CLI integration for non-visualize paths (MUST)
+
+Add `--sync` flag to `pactkit visualize` that forces codegraph sync even when `--lazy` skips graph regeneration. This covers the Hotfix use case where visualize is not run but codegraph still needs syncing.
+
+Also: expose `pactkit sync` as a standalone subcommand that only runs codegraph sync (no mermaid graphs). This is the minimal replacement for prompt instructions in Hotfix flow.
+
+### R4: Remove prompt instructions (MUST)
+
+Remove all "If `.codegraph/` exists, run `codegraph sync`" instructions from:
+- `commands.py:248` (Act Phase 4)
+- `commands.py:505` (Done Phase 2)
+- `workflows.py:614` (Hotfix Phase 3.6)
+
+Replace with a note that `pactkit visualize --lazy` and `pactkit sync` handle this automatically.
+
+### R5: Output feedback (SHOULD)
+
+When codegraph sync runs, print a brief status line: `"🔄 codegraph synced (N files updated)"` or `"codegraph: skipped (not installed)"`. Silent when `.codegraph/` doesn't exist (user hasn't opted in).
+
+## Acceptance Criteria
+
+### AC1: visualize --lazy triggers codegraph sync when source changed (R1)
+
+- **Given** project has `.codegraph/` directory and `codegraph` is on PATH
+- **When** `pactkit visualize --lazy` runs and source files have changed
+- **Then** `codegraph sync` is executed after mermaid graph generation, output includes sync status
+
+### AC2: visualize --lazy skips codegraph sync when no .codegraph/ (R1, R2)
+
+- **Given** project does NOT have `.codegraph/` directory
+- **When** `pactkit visualize --lazy` runs
+- **Then** no codegraph sync is attempted, no error is raised
+
+### AC3: codegraph_sync returns gracefully when binary missing (R2)
+
+- **Given** project has `.codegraph/` directory but `codegraph` is NOT on PATH
+- **When** `codegraph_sync(project_root)` is called
+- **Then** returns `(False, "codegraph not installed")` without raising
+
+### AC4: pactkit sync standalone command (R3)
+
+- **Given** project has `.codegraph/` and `codegraph` on PATH
+- **When** user runs `pactkit sync`
+- **Then** codegraph sync runs and reports status
+
+### AC5: prompt instructions removed (R4)
+
+- **Given** the updated codebase
+- **When** `grep -rn "codegraph sync" src/pactkit/prompts/` is run
+- **Then** zero matches for "If .codegraph/ exists, run codegraph sync" pattern (only informational references like skills.py docs remain)
+
+### AC6: output feedback on sync (R5)
+
+- **Given** project has `.codegraph/` and codegraph is on PATH
+- **When** `pactkit visualize --lazy` runs and triggers sync
+- **Then** stdout includes a status line like "🔄 codegraph synced" or "codegraph: skipped (not installed)"
+
+## Target Call Chain
+
+```
+CLI main() → args.command == "visualize"
+  → lazy_visualize.should_visualize()
+  → run_visualize_graphs() / run_visualize_single()
+  → [NEW] codegraph_sync(project_root)
+      → shutil.which("codegraph")
+      → subprocess.run(["codegraph", "sync", str(project_root)])
+
+CLI main() → args.command == "sync"  [NEW]
+  → codegraph_sync(project_root)
+```
+
+## Implementation Steps
+
+| Step | File | Action | Dependencies | Risk |
+|------|------|--------|-------------|------|
+| 1 | `src/pactkit/lazy_visualize.py` | Add `codegraph_sync(project_root) -> tuple[bool, str]` function | None | Low |
+| 2 | `src/pactkit/lazy_visualize.py` | Call `codegraph_sync()` at end of `run_visualize_graphs()` and `run_visualize_single()` | Step 1 | Low |
+| 3 | `src/pactkit/cli.py` | Add `sync` subcommand that calls `codegraph_sync()` | Step 1 | Low |
+| 4 | `src/pactkit/cli.py` | In `visualize` handler, call `codegraph_sync()` after lazy check passes | Step 1 | Low |
+| 5 | `src/pactkit/prompts/commands.py` | Remove "If .codegraph/ exists, run codegraph sync" from Act & Done prompts | None | Low |
+| 6 | `src/pactkit/prompts/workflows.py` | Remove Hotfix Phase 3.6 codegraph sync prompt instruction | None | Low |
+| 7 | `tests/unit/` | Unit tests for `codegraph_sync()` with mocked subprocess | Step 1 | Low |
+
+## Security Scope
+
+| Check | Applicable | Reason |
+|-------|------------|--------|
+| SEC-1 Input Validation | N/A | No user input — project_root from cwd |
+| SEC-2 Authentication | N/A | Local CLI tool |
+| SEC-3 Authorization | N/A | Local CLI tool |
+| SEC-4 Data Exposure | N/A | No secrets involved |
+| SEC-5 Injection | N/A | subprocess uses list args, not shell=True |
+| SEC-6 Dependencies | N/A | No new dependencies |
+| SEC-7 Cryptography | N/A | Not applicable |
+| SEC-8 Logging | N/A | No sensitive data logged |
+
+## Out of Scope
+
+- codegraph MCP server auto-start (out of scope — user manages MCP servers independently)
+- codegraph init auto-setup (already handled in Plan Phase 1 prompt, keep as-is)
+- Modifying `pactkit query` behavior (already works correctly)

pactkit-2.14.2/docs/specs/STORY-slim-127.md

@@ -0,0 +1,116 @@
+# STORY-slim-127: project CLAUDE.md managed-block update
+
+| Field | Value |
+|-------|-------|
+| ID | STORY-slim-127 |
+| Status | Done |
+| Priority | P1 |
+| Release | 2.15.0 |
+
+## Background
+
+`_generate_project_claude_md` currently uses `atomic_write` to fully overwrite `.claude/CLAUDE.md` on every `pactkit update`. This violates Architecture Principle §9 (Merge over Replace) — any content the user manually adds to CLAUDE.md is silently destroyed.
+
+The fix: adopt the same managed-block pattern used by `_upsert_venv_managed_block` in CLAUDE.local.md. PactKit owns a `<!-- pactkit:start -->` / `<!-- pactkit:end -->` region; user content outside that region is preserved.
+
+## Requirements
+
+### R1: Managed-block update for project CLAUDE.md (MUST)
+
+`_generate_project_claude_md` MUST use `<!-- pactkit:start -->` / `<!-- pactkit:end -->` markers. Only content between these markers is regenerated. User content outside is preserved.
+
+### R2: Migration from legacy full-replace (MUST)
+
+When CLAUDE.md exists but has no markers:
+- If content matches PactKit template (detected via `# {project_name} — Project Context` header) → wrap entire content in markers (safe migration).
+- If content has user modifications (no PactKit header) → append managed block at end of file, preserve all existing content.
+
+### R3: Fresh install (MUST)
+
+When CLAUDE.md does not exist, create it with the managed block including markers. Also include `@` import references outside the managed block so users can add content above or below.
+
+### R4: Preserve @import references position (MUST)
+
+The `@./docs/product/context.md` and `@./.claude/CLAUDE.local.md` references MUST be placed outside the managed block (after `<!-- pactkit:end -->`) so they are never accidentally removed if the user rearranges content.
+
+### R5: Codegraph section conditional inclusion (SHOULD)
+
+When `.codegraph/` exists, include the "Code Intelligence" section inside the managed block. When absent, omit it. This is dynamic per-deploy.
+
+## Acceptance Criteria
+
+### AC1: Existing user content preserved on update (R1)
+
+- **Given** `.claude/CLAUDE.md` exists with user content outside markers
+- **When** `pactkit update` runs
+- **Then** user content is unchanged; only content between `<!-- pactkit:start -->` and `<!-- pactkit:end -->` is regenerated
+
+### AC2: Legacy CLAUDE.md migrated with markers (R2)
+
+- **Given** `.claude/CLAUDE.md` exists with PactKit-generated content but no markers
+- **When** `pactkit update` runs
+- **Then** content is wrapped in markers; no content is lost
+
+### AC3: User-modified legacy CLAUDE.md preserved (R2)
+
+- **Given** `.claude/CLAUDE.md` exists with user-written content (no PactKit header)
+- **When** `pactkit update` runs
+- **Then** existing content is preserved, managed block is appended
+
+### AC4: Fresh install creates file with markers (R3)
+
+- **Given** `.claude/CLAUDE.md` does not exist
+- **When** `pactkit update` runs
+- **Then** file is created with `<!-- pactkit:start -->` ... `<!-- pactkit:end -->` and `@` imports after the end marker
+
+### AC5: @imports are outside managed block (R4)
+
+- **Given** any state of CLAUDE.md
+- **When** `pactkit update` runs
+- **Then** `@./docs/product/context.md` and `@./.claude/CLAUDE.local.md` appear after `<!-- pactkit:end -->`
+
+### AC6: Codegraph section appears when .codegraph/ exists (R5)
+
+- **Given** project has `.codegraph/` directory
+- **When** `pactkit update` runs
+- **Then** managed block includes "Code Intelligence (codegraph)" section
+
+## Target Call Chain
+
+```
+deployer.deploy() → _generate_project_claude_md(config)
+  → _build_managed_block(config, project_root)  [NEW]
+  → _upsert_claude_md_managed_block(claude_md_path, managed_block)  [NEW]
+      → if file missing: write fresh (markers + @imports)
+      → if has markers: regex replace between markers
+      → if legacy PactKit template: wrap in markers
+      → if user content: append managed block
+```
+
+## Implementation Steps
+
+| Step | File | Action | Dependencies | Risk |
+|------|------|--------|-------------|------|
+| 1 | `src/pactkit/generators/deployer.py` | Add `_CLAUDE_MD_START` / `_CLAUDE_MD_END` constants | None | Low |
+| 2 | `src/pactkit/generators/deployer.py` | Extract `_build_claude_md_managed_content()` from existing lines[] logic | None | Low |
+| 3 | `src/pactkit/generators/deployer.py` | Rewrite `_generate_project_claude_md` to use upsert pattern (detect markers, migrate, or fresh create) | Steps 1-2 | Medium |
+| 4 | `tests/unit/` | Unit tests for all 4 paths (fresh, has-markers, legacy-template, user-modified) | Step 3 | Low |
+
+## Security Scope
+
+| Check | Applicable | Reason |
+|-------|------------|--------|
+| SEC-1 Input Validation | N/A | No user input — paths from cwd |
+| SEC-2 Authentication | N/A | Local CLI tool |
+| SEC-3 Authorization | N/A | Local CLI tool |
+| SEC-4 Data Exposure | N/A | No secrets |
+| SEC-5 Injection | N/A | No shell/SQL |
+| SEC-6 Dependencies | N/A | No new deps |
+| SEC-7 Cryptography | N/A | Not applicable |
+| SEC-8 Logging | N/A | No sensitive data |
+
+## Out of Scope
+
+- Global `~/.claude/CLAUDE.md` — already has user-modified detection, not changing
+- `CLAUDE.local.md` — already uses managed-block pattern, not changing
+- Plugin/marketplace format CLAUDE.md — separate code path

{pactkit-2.14.1 → pactkit-2.14.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "pactkit"
-version = "2.14.1"
+version = "2.14.2"
 description = "Spec-driven agentic DevOps toolkit for AI coding assistants"
 readme = "README.md"
 license = "MIT"

{pactkit-2.14.1 → pactkit-2.14.2}/src/pactkit/__init__.py

@@ -1,3 +1,3 @@
 """PactKit - Spec-driven agentic DevOps toolkit."""
 
-__version__ = "2.14.1"
+__version__ = "2.14.2"

{pactkit-2.14.1 → pactkit-2.14.2}/src/pactkit/cli.py

@@ -268,6 +268,10 @@ def main():
     viz_parser.add_argument("--reverse", action="store_true", default=False, help="Reverse BFS: find callers of entry")
     viz_parser.add_argument("--depth", type=int, default=0, help="Limit traversal depth (0=unlimited)")
     viz_parser.add_argument("--max-nodes", type=int, default=0, help="Truncate graph to N nodes (0=unlimited)")
+    viz_parser.add_argument("--sync", action="store_true", help="Force codegraph sync even if graphs are skipped")
+
+    # pactkit sync (STORY-slim-126: codegraph sync code enforcement)
+    subparsers.add_parser("sync", help="Sync codegraph index")
 
     # pactkit garden (STORY-slim-070)
     garden_parser = subparsers.add_parser("garden", help="Codebase quality patrol")
@@ -565,12 +569,18 @@ def main():
     elif args.command == "visualize":
         from pathlib import Path
 
-        from pactkit.lazy_visualize import run_visualize_graphs, run_visualize_single, should_visualize
+        from pactkit.lazy_visualize import codegraph_sync, run_visualize_graphs, run_visualize_single, should_visualize
 
         project_root = Path.cwd()
         if args.lazy:
             should_run, reason = should_visualize(project_root, stack=args.stack)
             if not should_run:
+                if args.sync:
+                    synced, msg = codegraph_sync(project_root)
+                    if synced:
+                        print(f"🔄 {msg}")
+                    elif "skipped" not in msg:
+                        print(f"codegraph: {msg}")
                 print(reason)
                 raise SystemExit(0)
             print(f"Visualize needed: {reason}")
@@ -584,6 +594,18 @@ def main():
         else:
             run_visualize_graphs(project_root, focus=args.focus)
 
+    elif args.command == "sync":
+        from pathlib import Path
+
+        from pactkit.lazy_visualize import codegraph_sync
+
+        project_root = Path.cwd()
+        synced, msg = codegraph_sync(project_root)
+        if synced:
+            print(f"🔄 {msg}")
+        else:
+            print(f"codegraph: {msg}")
+
     elif args.command == "garden":
         from pathlib import Path
 

{pactkit-2.14.1 → pactkit-2.14.2}/src/pactkit/config.py

@@ -478,25 +478,9 @@ def auto_merge_config_file(path: Union[Path, str]) -> list[str]:
             for item in new_items:
                 added.append(f"{key}: {item}")
 
-    # --- Non-list sections: backfill missing with defaults (STORY-033, STORY-039) ---
-    defaults = get_default_config()
-    _BACKFILL_KEYS = (
-        "ci",
-        "issue_tracker",
-        "lint_blocking",
-        "auto_fix",
-        "venv",
-        "release",
-        "regression",
-        "check",
-        "done",
-        "e2e",  # STORY-slim-022
-        "visualize",  # STORY-slim-028
-    )
-    for key in _BACKFILL_KEYS:
-        if key not in user_data:
-            user_data[key] = defaults[key]
-            added.append(f"section: {key}")
+    # Non-list sections are NO LONGER backfilled (STORY-slim-126).
+    # Absent key = accept default. Users keep yaml minimal.
+    # Backfill only happens during `pactkit init` (scaffold).
 
     # STORY-slim-102: Remove stale version field from project yaml
     if "version" in user_data: