pactkit 2.9.13__tar.gz → 2.10.2__tar.gz

pactkit-2.10.2/.github/workflows/pactkit.yml

@@ -0,0 +1,31 @@
+name: PactKit CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+    branches: [main]
+
+jobs:
+  test:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Set up Python
+        uses: actions/setup-python@v5
+        with:
+          python-version: "3.11"
+
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -e .
+          pip install pytest ruff
+          pactkit init
+
+      - name: Lint
+        run: ruff check src/ tests/
+
+      - name: Test
+        run: pytest tests/ -v

{pactkit-2.9.13 → pactkit-2.10.2}/.gitignore

@@ -38,9 +38,17 @@ docs/architecture/snapshots/
 # Codex CLI local config
 .codex/
 
+# GitHub Copilot deployment artifacts (generated by pactkit update)
+.github/
+!.github/workflows/
+
 # Plugin deployment artifacts (generated by pactkit update)
 pactkit-plugin/
 
+# Office files
+~$*
+*.pptx
+
 # Secrets
 .pypirc
 

{pactkit-2.9.13 → pactkit-2.10.2}/.opencode/pactkit.yaml

@@ -2,11 +2,77 @@
 # Edit this file to customize which components are deployed.
 # Remove items from a list to disable them. Default: all enabled.
 
-version: "2.9.13"
+version: "2.10.2"
 stack: python
 root: .
 developer: "slim"
 
+# Agents — AI role definitions
+agents:
+  - code-explorer
+  - product-designer
+  - qa-engineer
+  - repo-maintainer
+  - security-auditor
+  - senior-developer
+  - system-architect
+  - system-medic
+  - visual-architect
+
+# Commands — PDCA playbooks
+commands:
+  - project-act
+  - project-check
+  - project-clarify
+  - project-design
+  - project-done
+  - project-hotfix
+  - project-init
+  - project-plan
+  - project-pr
+  - project-release
+  - project-sprint
+
+# Skills — tool scripts
+skills:
+  - pactkit-analyze
+  - pactkit-board
+  - pactkit-doctor
+  - pactkit-draw
+  - pactkit-release
+  - pactkit-review
+  - pactkit-scaffold
+  - pactkit-status
+  - pactkit-trace
+  - pactkit-visualize
+  - pactkit-garden
+  - project-act
+  - project-check
+  - project-clarify
+  - project-design
+  - project-done
+  - project-hotfix
+  - project-init
+  - project-plan
+  - project-pr
+  - project-release
+  - project-sprint
+  - pactkit-audit
+  - pactkit-report
+
+# Rules — constitution modules
+rules:
+  - 01-core-protocol
+  - 02-hierarchy-of-truth
+  - 03-file-atlas
+  - 04-routing-table
+  - 05-workflow-conventions
+  - 06-mcp-integration
+  - 07-shared-protocols
+  - 08-architecture-principles
+  - 09-sectional-write
+  - 11-pdca-nudge
+
 # CI/CD — set provider to github or gitlab to generate pipeline config
 ci:
   provider: github
@@ -46,6 +112,16 @@ regression:
 check:
   security_checklist: true
   security_scope_override: none
+  pactguard:
+    enabled: false
+    mode: all
+    ruleset: ""
+    blocking: false
+  observe:
+    enabled: false
+    sources: auto
+    max_console: 100
+    max_network: 200
 
 # Done — configure commit and lesson quality behavior
 done:

{pactkit-2.9.13 → pactkit-2.10.2}/CHANGELOG.md

@@ -4,6 +4,26 @@ All notable changes to PactKit will be documented in this file.
 
 Format follows [Keep a Changelog](https://keepachangelog.com/).
 
+## [2.10.2] - 2026-04-20
+
+### Added
+- **PDCA Nudge Protocol** (STORY-slim-098) — AI proactively recommends PDCA commands when free conversation yields actionable conclusions. Trigger matrix maps signals to commands; suppression rules prevent noise.
+
+### Fixed
+- **Shared Protocols Context.md reference** (STORY-slim-099) — Added missing `Act Phase 4` to the Context.md Canonical Format "Referenced by" line. Ensures context.md reflects Act progress for session continuity.
+- **Semantic version comparison** (HOTFIX-slim-099) — Version mismatch warning now uses tuple comparison instead of string equality, giving correct upgrade/downgrade direction.
+
+## [2.10.1] - 2026-04-16
+
+### Added
+- **Dual-dimension Harness Audit** (STORY-slim-097) — Audit now scores two dimensions: Config (project + global `~/.claude/` config, 50pts) and Code (tests, lint, complexity, git hygiene, 50pts). Score went from 52 to 97/100 for PactKit itself. JSON output includes `dimensions` breakdown field.
+- **Unified HTML Report Dashboard** (STORY-slim-094) — Single `report.html` with tab switching, D3 force-directed graph, harness score ring, layer bars, and hotspot panel. Self-contained offline HTML with inline D3.js.
+
+### Fixed
+- **Focus call graph empty output** (STORY-slim-095) — Fixed focus resolution to use `LANG_PROFILES[stack].source_dirs` instead of hardcoded `src/` prefix. Added subdirectory and single-root-module fallback.
+- **Report --all tab overload** (HOTFIX-slim-096) — Filtered unified dashboard to core PDCA graphs only (code_graph, class_graph, call_graph, system_design). Reduced from 9 tabs to 4.
+- **Mermaid `<br/>` tag rendering** (HOTFIX-slim-096) — Strip Mermaid `<br/>` tags from node labels before HTML escape, preventing `&lt;br/&gt;` raw text in SVG.
+
 ## [2.9.13] - 2026-04-15
 
 ### Fixed

{pactkit-2.9.13 → pactkit-2.10.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pactkit
-Version: 2.9.13
+Version: 2.10.2
 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

{pactkit-2.9.13 → pactkit-2.10.2}/docs/architecture/governance/archive/lessons_archive_202603.md

@@ -35,3 +35,14 @@
 | 2026-03-21 | Migrating ~25 deterministic rules from MD prompts to Python CLI subcommands (8 new modules: guards.py, id_generator.py, cleaners.py, regression.py, context_gen.py, validators.py, sec_scope.py, lazy_visualize.py) — the hardest part is backward compatibility: 2500+ pre-existing tests assert specific keywords in prompt text, so slimmed prompts must preserve every keyword even when replacing verbose blocks with single CLI invocations; build a keyword constraint map from failing tests incrementally rather than trying to predict all constraints up-front | STORY-slim-014 |
 | 2026-03-21 | Config resolution in CLI modules must use `find_pactkit_yaml(project_root)` → `load_config(yaml_path)` two-step pattern — `load_config(project_root)` alone fails because pactkit.yaml lives in `.claude/` subdirectory, not project root; lint_runner.py and test_mapper.py both reuse `LANG_PROFILES` from workflows.py as canonical source for stack-specific patterns (test_map_pattern, lint_command), keeping DRY with the CI pipeline profiles | STORY-slim-016 |
 | 2026-03-21 | Cross-flow integrity audits (comparing all PDCA commands end-to-end) catch gaps that per-command reviews miss — CLI subcommands with zero prompt references are dead code (`lint-context`, `lint-lessons`, `lint-testcase`); argparser flags without `deploy()` signature parity (`upgrade` missing `--agent`) fail silently; adding a new parameter to `deploy()` must use a default value to avoid breaking 24+ existing callers that don't pass it | BUG-slim-004 |
+| 2026-03-22 | Done Phase deterministic operations (lessons dedup, invariants regex, coverage parsing) are pure text/regex/subprocess with zero AI judgment — `lessons.py` uses Jaccard similarity on word sets (threshold < 0.5) for dedup, `invariants.py` uses `re.sub()` for test count replacement, `coverage_gate.py` uses list-form `subprocess.run()` for pytest-cov (SEC-7 safe); prompt delegation must preserve fallback instructions ("if pactkit X unavailable, fall back to manual") so LLM can still operate without CLI installed | STORY-slim-017 |
+| 2026-03-22 | Iterative cross-flow audits find deeper gaps each pass — BUG-slim-004 found 6 gaps, BUG-slim-005 found 5 more; dict keys defined but never consumed (`LANG_PROFILES.test_dir/package_file/e2e_test_pattern`) are invisible to grep-based prompt audits because they only exist in data structures; board update instructions must reference `{BOARD_CMD} update_task` script (not vague "mark as done") for deterministic execution; every CLI subcommand must have at least one prompt reference or it's dead code (`lint-testcase` had zero) | BUG-slim-005 |
+| 2026-03-21 | cli.py visualize --lazy was a decision-only tool that printed guidance but never called lazy_visualize.run_visualize_graphs(); always trace the execution path from CLI handler to actual side-effect to confirm end-to-end behavior | BUG-slim-006 |
+| 2026-03-21 | 5 test files independently hardcoded LANG_PROFILES required keys — define canonical key sets as frozenset in schemas.py (LANG_PROFILE_REQUIRED_KEYS) so all consumers import from one source; when adding test assertions on dict keys, always check schemas.py first | STORY-slim-018 |
+| 2026-03-23 | Monolithic prompt Phase 3.2 in commands.py caused AI thinking stalls; splitting into 4 sub-phases (3.2a-3.2d) with output checkpoints forces incremental generation and eliminates the bottleneck | src/pactkit/prompts/commands.py:COMMAND_PROMPTS[project-plan.md] |
+| 2026-03-23 | code-explorer maxTurns reduced from 50 to 15 in agents.py; Plan Phase 1 now requires bounded Explore prompts with target/scope/limit/output via Delegation Template in commands.py | src/pactkit/prompts/agents.py:code-explorer.maxTurns,src/pactkit/prompts/commands.py:Phase1.DelegationTemplate |
+| 2026-03-23 | Added 09-sectional-write as core rule in src/pactkit/prompts/rules.py RULES_MODULES. Threshold 300 lines based on P75-P90 analysis of specs/source/tests. Applies to any file type. | src/pactkit/prompts/rules.py:RULES_MODULES |
+| 2026-03-23 | When adding a new pactkit.yaml config section, 4 touch points needed: get_default_config, DEEP_MERGE_KEYS, validate_config, generate_default_yaml. Table format in prompts saves chars vs individual sections. | config.py:get_default_config |
+| 2026-03-24 | Plan Phase 3.2a scaffold-first: replaced AI freeform Write with {SCAFFOLD_CMD} create_spec + Read + Edit in src/pactkit/prompts/commands.py. Removed 3 inline format examples (~240 chars). Pre-existing tests (test_bug034, test_story055_commands, test_story_slim019) needed updates for new checkpoint wording. | src/pactkit/prompts/commands.py:Phase3.2a |
+| 2026-03-24 | spec_linter.py _check_acceptance_criteria: per-subsection GWT check must use raw_text (with code blocks) not stripped text, because specs legitimately wrap Gherkin in fenced blocks | spec_linter.py:_check_acceptance_criteria |
+| 2026-03-24 | Standalone scripts (visualize.py) need try/except ImportError guards for yaml.safe_load when reading pactkit.yaml — they cannot import from pactkit library | visualize.py:_load_scan_excludes |

pactkit-2.10.2/docs/architecture/governance/harness_audit.json

@@ -0,0 +1,197 @@
+{
+  "timestamp": "2026-04-16T12:48:17Z",
+  "commit": "517b0d0",
+  "story_id": "",
+  "score": 97,
+  "ready": true,
+  "weakest": null,
+  "layers": {
+    "H1": {
+      "level": 3,
+      "name": "Advanced"
+    },
+    "H2": {
+      "level": 3,
+      "name": "Advanced"
+    },
+    "H3": {
+      "level": 3,
+      "name": "Advanced"
+    },
+    "H4": {
+      "level": 2,
+      "name": "Structured"
+    },
+    "H5": {
+      "level": 3,
+      "name": "Advanced"
+    },
+    "H6": {
+      "level": 3,
+      "name": "Advanced"
+    },
+    "H7": {
+      "level": 3,
+      "name": "Advanced"
+    }
+  },
+  "dimensions": {
+    "config": {
+      "score": 50,
+      "passed": 13,
+      "total": 13
+    },
+    "code": {
+      "score": 47,
+      "passed": 16,
+      "total": 17
+    }
+  },
+  "hotspots": [
+    {
+      "file": ".github/skills/pactkit-visualize/scripts/visualize.py",
+      "score": 29,
+      "complexity_avg": 9.5,
+      "blast_pct": 0,
+      "fan_in": 0,
+      "function_count": 140,
+      "has_test": false,
+      "docstring_pct": 67,
+      "long_funcs": 19,
+      "deep_nesting": 34,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 19 long functions, 34 deeply nested"
+    },
+    {
+      "file": "src/pactkit/skills/visualize.py",
+      "score": 20,
+      "complexity_avg": 10.8,
+      "blast_pct": 0,
+      "fan_in": 0,
+      "function_count": 85,
+      "has_test": false,
+      "docstring_pct": 73,
+      "long_funcs": 14,
+      "deep_nesting": 18,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 14 long functions, 18 deeply nested"
+    },
+    {
+      "file": "src/pactkit/cli.py",
+      "score": 16,
+      "complexity_avg": 50.0,
+      "blast_pct": 2,
+      "fan_in": 1,
+      "function_count": 2,
+      "has_test": false,
+      "docstring_pct": 50,
+      "long_funcs": 1,
+      "deep_nesting": 1,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Split: extract high-complexity functions"
+    },
+    {
+      "file": "pactkit-plugin/skills/pactkit-visualize/scripts/visualize.py",
+      "score": 13,
+      "complexity_avg": 7.4,
+      "blast_pct": 0,
+      "fan_in": 0,
+      "function_count": 80,
+      "has_test": false,
+      "docstring_pct": 65,
+      "long_funcs": 7,
+      "deep_nesting": 13,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 7 long functions, 13 deeply nested"
+    },
+    {
+      "file": "src/pactkit/config.py",
+      "score": 9,
+      "complexity_avg": 15.6,
+      "blast_pct": 16,
+      "fan_in": 7,
+      "function_count": 12,
+      "has_test": true,
+      "docstring_pct": 100,
+      "long_funcs": 6,
+      "deep_nesting": 2,
+      "layer_violations": 0,
+      "action": "Refactor: 6 long functions, 2 deeply nested + Stabilize: 7 dependents"
+    },
+    {
+      "file": "src/pactkit/audit.py",
+      "score": 8,
+      "complexity_avg": 10.9,
+      "blast_pct": 2,
+      "fan_in": 1,
+      "function_count": 26,
+      "has_test": true,
+      "docstring_pct": 96,
+      "long_funcs": 5,
+      "deep_nesting": 6,
+      "layer_violations": 0,
+      "action": "Refactor: 5 long functions, 6 deeply nested + Decompose: 26 functions"
+    },
+    {
+      "file": "src/pactkit/skills/analyzers/go_analyzer.py",
+      "score": 8,
+      "complexity_avg": 9.5,
+      "blast_pct": 0,
+      "fan_in": 0,
+      "function_count": 11,
+      "has_test": false,
+      "docstring_pct": 36,
+      "long_funcs": 2,
+      "deep_nesting": 3,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 2 long functions, 3 deeply nested"
+    },
+    {
+      "file": "src/pactkit/skills/report.py",
+      "score": 7,
+      "complexity_avg": 9.1,
+      "blast_pct": 2,
+      "fan_in": 1,
+      "function_count": 11,
+      "has_test": false,
+      "docstring_pct": 100,
+      "long_funcs": 5,
+      "deep_nesting": 0,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 5 long functions"
+    },
+    {
+      "file": "src/pactkit/generators/deployer.py",
+      "score": 7,
+      "complexity_avg": 5.5,
+      "blast_pct": 5,
+      "fan_in": 2,
+      "function_count": 41,
+      "has_test": false,
+      "docstring_pct": 95,
+      "long_funcs": 8,
+      "deep_nesting": 0,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 8 long functions"
+    },
+    {
+      "file": "src/pactkit/skills/analyzers/java_analyzer.py",
+      "score": 7,
+      "complexity_avg": 7.1,
+      "blast_pct": 0,
+      "fan_in": 0,
+      "function_count": 11,
+      "has_test": false,
+      "docstring_pct": 36,
+      "long_funcs": 1,
+      "deep_nesting": 3,
+      "layer_violations": 0,
+      "action": "Add tests: no test file found + Refactor: 1 long functions, 3 deeply nested"
+    }
+  ],
+  "suggested_tasks": [],
+  "dependency_health": {
+    "vulns": -1,
+    "error": "pip-audit not installed"
+  }
+}

{pactkit-2.9.13 → pactkit-2.10.2}/docs/architecture/governance/lessons.md

@@ -2,17 +2,6 @@
 
 | Date | Lesson | Context |
 |------|--------|---------|
-| 2026-03-22 | Done Phase deterministic operations (lessons dedup, invariants regex, coverage parsing) are pure text/regex/subprocess with zero AI judgment — `lessons.py` uses Jaccard similarity on word sets (threshold < 0.5) for dedup, `invariants.py` uses `re.sub()` for test count replacement, `coverage_gate.py` uses list-form `subprocess.run()` for pytest-cov (SEC-7 safe); prompt delegation must preserve fallback instructions ("if pactkit X unavailable, fall back to manual") so LLM can still operate without CLI installed | STORY-slim-017 |
-| 2026-03-22 | Iterative cross-flow audits find deeper gaps each pass — BUG-slim-004 found 6 gaps, BUG-slim-005 found 5 more; dict keys defined but never consumed (`LANG_PROFILES.test_dir/package_file/e2e_test_pattern`) are invisible to grep-based prompt audits because they only exist in data structures; board update instructions must reference `{BOARD_CMD} update_task` script (not vague "mark as done") for deterministic execution; every CLI subcommand must have at least one prompt reference or it's dead code (`lint-testcase` had zero) | BUG-slim-005 |
-| 2026-03-21 | cli.py visualize --lazy was a decision-only tool that printed guidance but never called lazy_visualize.run_visualize_graphs(); always trace the execution path from CLI handler to actual side-effect to confirm end-to-end behavior | BUG-slim-006 |
-| 2026-03-21 | 5 test files independently hardcoded LANG_PROFILES required keys — define canonical key sets as frozenset in schemas.py (LANG_PROFILE_REQUIRED_KEYS) so all consumers import from one source; when adding test assertions on dict keys, always check schemas.py first | STORY-slim-018 |
-| 2026-03-23 | Monolithic prompt Phase 3.2 in commands.py caused AI thinking stalls; splitting into 4 sub-phases (3.2a-3.2d) with output checkpoints forces incremental generation and eliminates the bottleneck | src/pactkit/prompts/commands.py:COMMAND_PROMPTS[project-plan.md] |
-| 2026-03-23 | code-explorer maxTurns reduced from 50 to 15 in agents.py; Plan Phase 1 now requires bounded Explore prompts with target/scope/limit/output via Delegation Template in commands.py | src/pactkit/prompts/agents.py:code-explorer.maxTurns,src/pactkit/prompts/commands.py:Phase1.DelegationTemplate |
-| 2026-03-23 | Added 09-sectional-write as core rule in src/pactkit/prompts/rules.py RULES_MODULES. Threshold 300 lines based on P75-P90 analysis of specs/source/tests. Applies to any file type. | src/pactkit/prompts/rules.py:RULES_MODULES |
-| 2026-03-23 | When adding a new pactkit.yaml config section, 4 touch points needed: get_default_config, DEEP_MERGE_KEYS, validate_config, generate_default_yaml. Table format in prompts saves chars vs individual sections. | config.py:get_default_config |
-| 2026-03-24 | Plan Phase 3.2a scaffold-first: replaced AI freeform Write with {SCAFFOLD_CMD} create_spec + Read + Edit in src/pactkit/prompts/commands.py. Removed 3 inline format examples (~240 chars). Pre-existing tests (test_bug034, test_story055_commands, test_story_slim019) needed updates for new checkpoint wording. | src/pactkit/prompts/commands.py:Phase3.2a |
-| 2026-03-24 | spec_linter.py _check_acceptance_criteria: per-subsection GWT check must use raw_text (with code blocks) not stripped text, because specs legitimately wrap Gherkin in fenced blocks | spec_linter.py:_check_acceptance_criteria |
-| 2026-03-24 | Standalone scripts (visualize.py) need try/except ImportError guards for yaml.safe_load when reading pactkit.yaml — they cannot import from pactkit library | visualize.py:_load_scan_excludes |
 | 2026-03-24 | Inline data in standalone scripts must have canonical-source comments pointing to the library module (e.g. _STACK_MARKERS → cleaners.py, _LANG_FILE_EXT → workflows.py) per Architecture Principle 1 | visualize.py:_STACK_MARKERS |
 | 2026-03-24 | Worktree isolation diverges from working-tree: verify visualize.py _scan_files() signature is preserved across stories to avoid breaking callers | visualize.py:_scan_files |
 | 2026-03-24 | Extracting _detect_stack() from _detect_file_ext() enables both file discovery and test mapping to share stack detection — DRY refactoring | visualize.py:_detect_stack |
@@ -53,3 +42,14 @@
 | 2026-04-02 | validate_deployed_content() guard caught real adapter bugs (missing _replace_slash_commands in copilot agents, missing _replace_cli_with_scripts in codex rules) — static analysis on deploy output is high-value for multi-adapter systems | deploy_base.py:validate_deployed_content |
 | 2026-04-02 | Borrowed Claude Code P3 (graduated safety language) + P6 (NO_TOOLS) patterns; applied to 01-core-protocol.md, 02-hierarchy-of-truth.md, 10-safety.md, project-act.md, project-check.md | src/pactkit/prompts/rules.py:Signal Strength Convention |
 | 2026-04-15 | pyproject.toml [project].dependencies must only include packages needed for core CLI startup (pyyaml); adapter and tree-sitter packages go in [project.optional-dependencies] with extras (opencode, codex, visualize, all) | pyproject.toml:[project.optional-dependencies] |
+| 2026-04-15 | _deploy_claude_md must read-before-write: user-modified global CLAUDE.md was silently destroyed on every pactkit init/update. Project-level _generate_project_claude_md had 3 protection layers; global had zero. | deployer.py:_deploy_claude_md |
+| 2026-04-16 | TreeSitterAnalyzer._compute_complexity() can serve as a universal base for all tree-sitter languages; only PythonAnalyzer needs a separate implementation using stdlib ast | analyzers/__init__.py:_compute_complexity |
+| 2026-04-16 | Parsing Mermaid .mmd files to JSON (nodes+edges) is straightforward with regex; the D3 force-directed HTML template works well as a single-file self-contained skill | report.py:_parse_mmd,_render_html |
+| 2026-04-16 | D3 force graph selectNode() computes blast radius via in-browser BFS — no backend needed; folder-based coloring via d3.scaleOrdinal on path prefix gives CodeFlow-quality visuals | src/pactkit/skills/report.py:selectNode |
+| 2026-04-16 | H1-H7 harness audit aggregator pattern: each _check_hN() function is self-contained with its own L0-L3 criteria, making it easy to add new checks per layer without touching scoring logic | src/pactkit/audit.py:_check_h1 |
+| 2026-04-16 | File-level hotspot aggregation (complexity_avg × blast_pct × fan_in) reduces 141 per-function findings to ~10 actionable items; _suggest_action() maps dominant signal to Split/Stabilize/Isolate/Decompose verb | src/pactkit/audit.py:_compute_hotspots,_suggest_action |
+| 2026-04-16 | Weighted hotspot formula (complexity 25% + docstring 15% + smells 15% + layers 10% + test 20% + blast 15%) gives meaningful scores across different project profiles; _generate_suggested_tasks auto-scaffolds BUG/HOTFIX specs with Done-completed filter for idempotency | src/pactkit/audit.py:_compute_hotspots,_generate_suggested_tasks |
+| 2026-04-16 | Dual-dimension audit (config+code 50/50) gives accurate project health picture; scanning ~/.claude/ global config catches harness setup invisible at project level | audit.py:_all_config_dirs |
+| 2026-04-17 | Dual-anchor pattern: declare principle in RULES_MODULES['core'] (01-core-protocol.md), detail in dedicated module (11-pdca-nudge.md). Prevents attention dilution when rules/ has 10+ files — Core Protocol acts as behavioral constitution anchor. | src/pactkit/prompts/rules.py:RULES_MODULES |
+| 2026-04-18 | Version comparison guards must use semantic (tuple) comparison, not string equality — string != treats 2.10.1 != 2.5.0 without direction, giving wrong upgrade advice | guards.py:check_version_mismatch |
+| 2026-04-20 | Plugin deployer (_deploy_commands in deployer.py:952-954) uses _rewrite_skills_prefix instead of _render_prompt, keeping raw placeholders like {BOARD_CMD} — not a deploy drift bug but by-design template behavior | src/pactkit/generators/deployer.py:_deploy_commands |

{pactkit-2.9.13 → pactkit-2.10.2}/docs/architecture/governance/rules.md

@@ -15,7 +15,7 @@
 
 ## Invariants
 
-1. All 3631+ tests must pass before any commit to `main`.
+1. All 3752+ 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-2.10.2/docs/product/archive/archive_202604.md

@@ -0,0 +1,207 @@
+
+### [STORY-slim-075] Prompt engineering quality: graduated safety language, tool call guidance, routing disambiguation, lessons rotation
+> Spec: docs/specs/STORY-slim-075.md
+
+- [x] Audit safety language,Add parallel/serial tool guidance,Add routing When NOT to use,Implement lessons rotation,Add prompt quality tests
+
+### [STORY-slim-076] Multi-stack visualize: class mode + multi-language file scanning
+> Spec: docs/specs/STORY-slim-076.md
+
+- [x] R1: _detect_stacks multi-stack detection
+- [x] R2: Multi-stack file scanning
+- [x] R3: extract_classes ABC
+- [x] R4: _build_class_graph refactor
+- [x] R5: _select_analyzers
+- [x] R6: Backward compat tests
+
+### [STORY-slim-077] Monorepo stack detection + redetect-stack CLI
+> Spec: docs/specs/STORY-slim-077.md
+
+- [x] Extend detect_stacks depth-1,Add redetect-stack CLI,Update init flow,Tests
+
+### [STORY-slim-078] Multi-language module resolution for file-mode dependency graph
+> Spec: docs/specs/STORY-slim-078.md
+
+- [x] Per-lang module keys,Per-lang import normalization,Multi-analyzer file graph,Fix src-strip,Go module prefix,Tests
+
+### [STORY-slim-079] TS/JS path alias resolution for file-mode dependency graph
+> Spec: docs/specs/STORY-slim-079.md
+
+- [x] Read tsconfig paths,Resolve alias imports,Wildcard patterns,Tests
+
+### [STORY-slim-082] Sync prompt templates for --mode module and --focus scoping
+> Spec: docs/specs/STORY-slim-082.md
+
+- [x] Update SKILL_VISUALIZE_MD
+- [x] Update Visual First rule
+- [x] Update release snapshot
+- [x] Update init Phase 3
+- [x] Deploy and verify
+
+### [STORY-slim-081] Two-tier module graph with scoped focus for large codebases
+> Spec: docs/specs/STORY-slim-081.md
+
+- [x] R1: _detect_modules boundary detection,R2: _build_module_graph with weighted edges,R3: Auto-degradation when files > MAX_SCAN_FILES,R4: Scoped focus scan (resolve module → directory),R5: .tsx/.jsx regression test,R6: Backward compat for small projects
+
+### [STORY-slim-080] Deep monorepo scanning: nearest-ancestor config discovery for all analyzers
+> Spec: docs/specs/STORY-slim-080.md
+
+- [x] Extend _detect_stacks depth,TS nearest-ancestor tsconfig,Go nearest-ancestor go.mod,Plumb consumer_path,Tests
+
+### [STORY-slim-083] Copilot deployer adapter package (pactkit-copilot)
+> Spec: docs/specs/STORY-slim-083.md
+
+- [x] CopilotDeployer class + entry_point registration + deploy to .github/ + OCP fix for rules_import_style dispatch
+
+### [STORY-slim-084] Adapter deploy-output validation guard
+> Spec: docs/specs/STORY-slim-084.md
+
+- [x] R2: has_pactkit_cli field,R1: validate_deployed_content(),R3: Core deploy integration,R6: Core unit tests,R4: Copilot integration test,R5: Codex integration test
+
+### [STORY-slim-086] Prompt Writing Quality — Signal Strength, Consequence Language, NO_TOOLS Mode
+> Spec: docs/specs/STORY-slim-086.md
+
+- [x] R1: Signal Strength Convention 追加到 01-core-protocol.md
+- [x] R2: 校准现有规则文件信号词 (01, 02, 10)
+- [x] R3: 7 条关键禁止规则添加后果语言
+- [x] R4: project-check.md 添加 NO_TOOLS restriction
+
+## ✅ Done
+
+- **STORY-slim-059**: Remove dead codex profile and slim down core package
+- **STORY-slim-058**: Extract pactkit-opencode as independent adapter package
+- **STORY-slim-057**: Refactor deployer.py: extract DeployerProtocol and DeployerBase
+
+### [HOTFIX-slim-085] Add Duplication Audit to Plan phase
+> Spec: docs/specs/HOTFIX-slim-085.md
+
+- [x] system-architect agent + Plan playbook
+
+### [STORY-slim-088] Slim dependencies and robust CLI fallback
+> Spec: docs/specs/STORY-slim-088.md
+
+- [x] Move adapters+tree-sitter to optional-dependencies
+- [x] Add python3 -m fallback for spec-lint in playbooks
+- [x] Fix add_story call signature in Plan playbook
+
+### [HOTFIX-slim-087] guard 添加 -C 参数
+> Spec: docs/specs/HOTFIX-slim-087.md
+
+- [x] Fix guard CWD false alarm in subagent
+
+### [BUG-slim-089] Global CLAUDE.md overwritten on every deploy
+> Spec: docs/specs/BUG-slim-089.md
+
+- [x] Add _is_pactkit_managed_global_md helper
+- [x] Refactor _deploy_claude_md with read-before-write guard
+- [x] Add unit tests for AC1-AC5
+
+### [STORY-slim-090] Interactive HTML Report Skill (D3 Force Graph)
+> Spec: docs/specs/STORY-slim-090.md
+
+- [x] Implement MMD parser (graph TD + classDiagram)
+- [x] Create D3 HTML template with force simulation
+- [x] Implement render_html with zoom/pan/drag/hover/search/theme
+- [x] Add overlay support (blast radius, complexity, layers)
+- [x] Create skill entry point + CLI
+- [x] Write unit tests
+
+### [STORY-slim-089] Enterprise Code Analysis: Blast Radius, Cyclomatic Complexity, Layer Violations
+> Spec: docs/specs/STORY-slim-089.md
+
+- [x] Add complexity counting to LanguageAnalyzer interface
+- [x] Implement Python complexity in PythonAnalyzer
+- [x] Implement TS/Go/Java complexity in TreeSitterAnalyzer subclasses
+- [x] Add blast_radius() with bidirectional BFS
+- [x] Add complexity() report function
+- [x] Add layers() with configurable layer model
+- [x] Add CLI subcommands (blast_radius, complexity, layers)
+- [x] Write unit tests for all 3 features
+
+### [STORY-slim-091] Harness Audit Skill: H1-H7 AI Readiness Assessment
+> Spec: docs/specs/STORY-slim-091.md
+
+- [x] Implement H1-H7 layer checks
+- [x] Implement scoring model (sum/21 + min≥L1)
+- [x] Implement findings aggregator
+- [x] Implement insights (fan-in, blast top10, circular deps, god objects)
+- [x] Register CLI subcommand (audit --json/--layer/--append)
+- [x] Integrate into Done Phase 3.x
+- [x] Integrate into Plan Phase 1 (optional read)
+- [x] Write unit tests
+
+### [STORY-slim-092] Audit Findings: File-Level Hotspot Aggregation
+> Spec: docs/specs/STORY-slim-092.md
+
+- [x] Add _compute_hotspots() file-level aggregation
+- [x] Add _suggest_action() actionable suggestions
+- [x] Slim _write_audit_json() to scorecard + hotspots only
+- [x] Add --verbose flag for full details
+- [x] Update tests for hotspot + slim JSON
+
+### [STORY-slim-093] Audit: Multi-Signal Hotspots + Suggested Tasks Generation
+> Spec: docs/specs/STORY-slim-093.md
+
+- [x] Add test coverage signal (_check_test_coverage)
+- [x] Add docstring coverage signal (_check_docstring_coverage)
+- [x] Add code smell signal (_check_code_smells)
+- [x] Add layer violation signal (_count_layer_violations)
+- [x] Add dependency health check (_check_dependency_health)
+- [x] Update hotspot score formula (weighted multi-dimensional)
+- [x] Update action suggestions (7-level priority)
+- [x] Add suggested tasks generation (_generate_suggested_tasks)
+- [x] Update JSON schema (suggested_tasks + dependency_health)
+- [x] Write unit tests for all new signals + tasks
+
+### [STORY-slim-097] Dual-dimension Harness Audit: Config + Code
+> Spec: docs/specs/STORY-slim-097.md
+
+- [x] R1: _all_config_dirs helper
+- [x] R2: Code dimension checks
+- [x] R3: Dual scoring formula
+- [x] R4: Layer level both dimensions
+- [x] R5: JSON dimensions breakdown
+- [x] R6: Global config scan
+
+### [HOTFIX-slim-096] Report --all filter core graphs only
+> Spec: docs/specs/HOTFIX-slim-096.md
+
+- [x] Filter unified dashboard to core PDCA graphs, exclude focus/reverse/workflow
+
+### [STORY-slim-094] Unified HTML Report — single aggregated dashboard
+> Spec: docs/specs/STORY-slim-094.md
+
+- [x] R1: Unified single HTML with Tab switching
+- [x] R2: Left panel audit dashboard
+- [x] R3: Right panel hotspot details
+- [x] R4: Single file backward compat
+- [x] R5: Self-contained offline HTML
+- [x] R6: Overlay auto-detect
+- [x] R7: CLI pactkit report subcommand
+
+### [STORY-slim-095] Fix visualize focus call graph + report empty tab
+> Spec: docs/specs/STORY-slim-095.md
+
+- [x] R1: Fix focus filter in _build_call_graph
+- [x] R2: Report empty Tab placeholder
+- [x] R3: Skip empty .mmd in unified report
+
+### [STORY-slim-098] PDCA Nudge Protocol — AI主动推荐PDCA命令
+> Spec: docs/specs/STORY-slim-098.md
+
+- [x] 创建11-pdca-nudge.md规则模板
+- [x] 在VALID_RULES注册新规则
+- [x] 添加部署测试
+- [x] 运行deploy验证
+
+### [HOTFIX-slim-099] Fix version mismatch guard semantic comparison
+> Spec: docs/specs/HOTFIX-slim-099.md
+
+- [x] Use packaging.version for directional version comparison in guards.py
+
+### [STORY-slim-099] Add Act Phase reference to Shared Protocols Context.md section
+> Spec: docs/specs/STORY-slim-099.md
+
+- [x] Update Shared Protocols Referenced-by line
+- [x] Redeploy via pactkit update
+- [x] Verify deployed plugin sync