@sienklogic/plan-build-run 2.0.0 → 2.0.2

package/CHANGELOG.md

@@ -1,56 +1,56 @@
-# Changelog
-
-All notable changes to Plan-Build-Run will be documented in this file.
-
-The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
-and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
-
-## [Unreleased]
-
-## [2.0.0] - 2026-02-17
-
-### Added
-- Token-saving CLI: 6 new commands in `towline-tools.js` — `frontmatter`, `must-haves`, `phase-info`, `state update`, `roadmap update-status`, `roadmap update-plans`
-- Companion web dashboard (Express 5.x, EJS, Pico.css v2, HTMX 2.0) with overview, phase detail, roadmap, todos, and SSE live updates
-- `/dev:import` skill — Import external plan documents into Towline format
-- `/dev:note` skill — Zero-friction idea capture with promote-to-todo support
-- `/dev:setup` skill — Interactive onboarding wizard for new installations
-- `/dev:explore` skill — Socratic conversation for idea exploration
-- `/dev:continue` skill — Execute the next logical step automatically
-- `/dev:health` skill — Validate `.planning/` directory integrity
-- General agent — Lightweight Towline-aware agent for ad-hoc tasks
-- `/dev:build <N> --team` variant — Agent Teams for complex inter-agent coordination
-- `/dev:review <N> --auto-fix` variant — Auto-diagnose and fix verification failures
-- Hook spawn tests for all lifecycle hooks
-- Iterative retrieval protocol for researcher agent
-- Behavioral contexts for agent prompt refinement
-- Published to npm with OIDC trusted publishing
-
-### Changed
-- Skill count increased from 15 to 21
-- Agent count increased from 9 to 10 (added General agent)
-- Hook scripts consolidated: Write/Edit dispatch reduced from 4 spawns to 2
-- All hook scripts now use `logHook()` from `hook-logger.js` for unified logging
-- Agents reference new CLI tooling shortcuts instead of manual YAML parsing
-- README rewritten with badges, comparison table, and acknowledgments
-- Package tarball trimmed from 1.9MB to 305KB via explicit `files` field
-
-### Fixed
-- Windows CI: `parseMustHaves` now trims CRLF line endings
-- Context budget: main orchestrator no longer reads agent definitions (saves ~15% context)
-- Hook logger rotation: `.hook-log` now caps at 200 entries with JSONL format
-- Status line ANSI rendering on Windows terminals
-
-## [1.0.0] - 2025-02-07
-
-### Added
-- Initial release of Plan-Build-Run plugin for Claude Code
-- 15 skills: begin, plan, build, review, discuss, quick, debug, status, pause, resume, milestone, scan, todo, config, help
-- 9 specialized agents: researcher, planner, plan-checker, executor, verifier, integration-checker, debugger, codebase-mapper, synthesizer
-- Hook-enforced quality gates: commit validation, plan format checking, session state injection, pre-compact preservation
-- Wave-based parallel execution via Task() subagents
-- Goal-backward verification at phase and milestone levels
-- Persistent file-based state management (.planning/ directory)
-- Configurable workflow: depth, models, gates, parallelization, git branching
-- Cross-platform Node.js hook scripts (Windows + macOS + Linux)
-- Plugin distribution via npm
+# Changelog
+
+All notable changes to Plan-Build-Run will be documented in this file.
+
+The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.1.0/),
+and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
+
+## [Unreleased]
+
+## [2.0.0] - 2026-02-17
+
+### Added
+- Token-saving CLI: 6 new commands in `towline-tools.js` — `frontmatter`, `must-haves`, `phase-info`, `state update`, `roadmap update-status`, `roadmap update-plans`
+- Companion web dashboard (Express 5.x, EJS, Pico.css v2, HTMX 2.0) with overview, phase detail, roadmap, todos, and SSE live updates
+- `/dev:import` skill — Import external plan documents into Towline format
+- `/dev:note` skill — Zero-friction idea capture with promote-to-todo support
+- `/dev:setup` skill — Interactive onboarding wizard for new installations
+- `/dev:explore` skill — Socratic conversation for idea exploration
+- `/dev:continue` skill — Execute the next logical step automatically
+- `/dev:health` skill — Validate `.planning/` directory integrity
+- General agent — Lightweight Towline-aware agent for ad-hoc tasks
+- `/dev:build <N> --team` variant — Agent Teams for complex inter-agent coordination
+- `/dev:review <N> --auto-fix` variant — Auto-diagnose and fix verification failures
+- Hook spawn tests for all lifecycle hooks
+- Iterative retrieval protocol for researcher agent
+- Behavioral contexts for agent prompt refinement
+- Published to npm with OIDC trusted publishing
+
+### Changed
+- Skill count increased from 15 to 21
+- Agent count increased from 9 to 10 (added General agent)
+- Hook scripts consolidated: Write/Edit dispatch reduced from 4 spawns to 2
+- All hook scripts now use `logHook()` from `hook-logger.js` for unified logging
+- Agents reference new CLI tooling shortcuts instead of manual YAML parsing
+- README rewritten with badges, comparison table, and acknowledgments
+- Package tarball trimmed from 1.9MB to 305KB via explicit `files` field
+
+### Fixed
+- Windows CI: `parseMustHaves` now trims CRLF line endings
+- Context budget: main orchestrator no longer reads agent definitions (saves ~15% context)
+- Hook logger rotation: `.hook-log` now caps at 200 entries with JSONL format
+- Status line ANSI rendering on Windows terminals
+
+## [1.0.0] - 2025-02-07
+
+### Added
+- Initial release of Plan-Build-Run plugin for Claude Code
+- 15 skills: begin, plan, build, review, discuss, quick, debug, status, pause, resume, milestone, scan, todo, config, help
+- 9 specialized agents: researcher, planner, plan-checker, executor, verifier, integration-checker, debugger, codebase-mapper, synthesizer
+- Hook-enforced quality gates: commit validation, plan format checking, session state injection, pre-compact preservation
+- Wave-based parallel execution via Task() subagents
+- Goal-backward verification at phase and milestone levels
+- Persistent file-based state management (.planning/ directory)
+- Configurable workflow: depth, models, gates, parallelization, git branching
+- Cross-platform Node.js hook scripts (Windows + macOS + Linux)
+- Plugin distribution via npm

package/CLAUDE.md

@@ -1,149 +1,149 @@
-# CLAUDE.md
-
-This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
-
-## What This Is
-
-Plan-Build-Run is a **Claude Code plugin** that provides a structured development workflow. It solves context rot — quality degradation as Claude's context window fills up — through disciplined subagent delegation, file-based state, and goal-backward verification. Users invoke `/pbr:*` slash commands (skills) that orchestrate specialized agents via `Task()`.
-
-## Critical Rules
-
-- **NEVER add AI co-author lines** to git commits or PRs. No `Co-Authored-By: Claude` or similar. Only add co-author lines referencing actual human contributors.
-- **NEVER inline agent definitions** into skill prompts. Use `subagent_type: "pbr:{name}"` — Claude Code auto-loads agent definitions from `agents/`. Reading agent `.md` files wastes main context.
-
-## Commands
-
-```bash
-npm test                                    # Run all Jest tests (~780 tests, 36 suites)
-npm run lint                                # ESLint on plugins/pbr/scripts/ and tests/
-npm run validate                            # Validate plugin directory structure
-npx jest tests/validate-commit.test.js      # Run a single test file
-npx jest --coverage                         # Run with coverage report
-```
-
-Coverage thresholds (enforced in `package.json`): 65% statements, 58% branches, 70% functions, 65% lines.
-
-Dashboard (separate dependency tree):
-```bash
-npm run dashboard:install                   # One-time install of dashboard deps
-npm run dashboard -- --dir /path/to/project # Launch dashboard for a project
-```
-
-Load the plugin locally for manual testing:
-```bash
-claude --plugin-dir .
-```
-
-CI runs on Node 18/20/22 across Windows, macOS, and Linux. All three platforms must pass.
-
-## Architecture
-
-All plugin code lives under `plugins/pbr/`. Three layers:
-
-### Skills (`skills/{name}/SKILL.md`)
-Markdown files with YAML frontmatter defining slash commands (`/pbr:begin`, `/pbr:plan`, etc.). Each SKILL.md is a complete prompt that tells the orchestrator what to do. Skills read state, interact with the user, and spawn agents.
-
-21 skills: begin, build, config, continue, debug, discuss, explore, health, help, import, milestone, note, pause, plan, quick, resume, review, scan, setup, status, todo.
-
-### Agents (`agents/{name}.md`)
-Markdown files with YAML frontmatter defining specialized subagent prompts. Agents run in fresh `Task()` contexts with clean 200k token windows. Spawned via `subagent_type: "pbr:{name}"` — auto-loaded by Claude Code.
-
-10 agents: researcher, planner, plan-checker, executor, verifier, integration-checker, debugger, codebase-mapper, synthesizer, general.
-
-### Scripts (`scripts/*.js`)
-28 Node.js hook scripts that fire on Claude Code lifecycle events. Configured in `hooks/hooks.json`. All use CommonJS, must be cross-platform (`path.join()`, not hardcoded separators), and log via `logHook()` from `hook-logger.js`.
-
-**Dispatch pattern**: Several hooks use dispatch scripts that fan out to sub-scripts based on the file being written/read:
-
-| Hook Event | Entry Script | Delegates To |
-|------------|-------------|-------------|
-| SessionStart | progress-tracker.js | — (injects project state) |
-| PostToolUse (Write\|Edit) | post-write-dispatch.js | check-plan-format.js, check-roadmap-sync.js, check-state-sync.js |
-| PostToolUse (Write\|Edit) | post-write-quality.js | check-doc-sprawl.js, check-skill-workflow.js |
-| PostToolUse (Task) | check-subagent-output.js | — (validates agent output) |
-| PostToolUse (Write\|Edit) | suggest-compact.js | — (context budget warnings) |
-| PostToolUse (Read) | track-context-budget.js | — (tracks reads for budget) |
-| PostToolUseFailure | log-tool-failure.js | — (logs failures) |
-| PreToolUse (Bash) | pre-bash-dispatch.js | validate-commit.js, check-dangerous-commands.js, check-phase-boundary.js |
-| PreToolUse (Write\|Edit) | pre-write-dispatch.js | — (write guards) |
-| PreCompact | context-budget-check.js | — (preserves STATE.md) |
-| Stop | auto-continue.js | — (chains next command) |
-| SubagentStart/Stop | log-subagent.js | — (tracks lifecycle) |
-| SubagentStop | event-handler.js | — (auto-verification trigger) |
-| TaskCompleted | task-completed.js | — (processes task completion) |
-| SessionEnd | session-cleanup.js | — (cleanup) |
-
-**Hook exit codes**: 0 = success, 2 = block (PreToolUse hooks that reject a tool call).
-
-**`${CLAUDE_PLUGIN_ROOT}`**: Used in hooks.json to reference script paths. Claude Code expands this internally — works on all platforms without shell expansion.
-
-### Supporting directories
-
-- **`references/`** — Shared reference docs loaded by skills (plan format, commit conventions, UI formatting, deviation rules)
-- **`templates/`** — EJS-style `.tmpl` files for generated markdown (VERIFICATION.md, SUMMARY.md, etc.)
-- **`commands/`** — Command registration files (one `.md` per command mapping to its skill)
-- **`skills/shared/`** — 14 shared skill fragments extracted from repeated patterns across skills (config-loading, digest-select, revision-loop, context-loader-task, universal-anti-patterns, phase-argument-parsing, gate-prompts, state-loading, state-update, commit-planning-docs, context-budget, domain-probes, progress-display, error-reporting)
-
-## Key Conventions
-
-**Commit format**: `{type}({scope}): {description}` — enforced by PreToolUse hook. Types: feat, fix, refactor, test, docs, chore, wip. Scopes: `{NN}-{MM}` (phase-plan), `quick-{NNN}`, `planning`, `tools`, or descriptive word.
-
-**Skill frontmatter** (SKILL.md):
-```yaml
----
-name: skill-name
-description: "What this skill does"
-allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Task
-argument-hint: "<N> [--flag]"
----
-```
-
-**Agent frontmatter** ({name}.md):
-```yaml
----
-name: agent-name
-description: "What this agent does"
-model: sonnet|inherit|haiku
-memory: none|user|project
-tools:
-  - Read
-  - Write
-  - Bash
----
-```
-
-## Data Flow
-
-Skills and agents communicate through files on disk, not messages:
-
-```
-.planning/STATE.md      ← source of truth for current position
-.planning/ROADMAP.md    ← phase structure, goals, dependencies
-.planning/config.json   ← workflow settings (~62 properties across 12 top-level keys)
-.planning/phases/NN-slug/
-  PLAN.md               ← written by planner, read by executor
-  SUMMARY.md            ← written by executor, read by orchestrator
-  VERIFICATION.md       ← written by verifier, read by review skill
-```
-
-The orchestrator stays lean (~15% context) by delegating heavy work to agents. Each agent gets a fresh context window.
-
-**Utility library**: `pbr-tools.js` is a shared Node.js library (stateLoad, configLoad, frontmatterParse, mustHavesCollect, etc.) used by multiple hook scripts. It provides CLI subcommands that agents call to avoid wasting tokens on file parsing.
-
-## Testing
-
-Tests live in `tests/` using Jest. Test files mirror script names: `validate-commit.test.js` tests `validate-commit.js`.
-
-**Fixture project**: `tests/fixtures/fake-project/.planning/` provides read-only fixture data for tests.
-
-**Mutation tests**: Use `fs.mkdtempSync()` to create temporary directories — never mutate the fixture project.
-
-**`pbr-tools.js` tests** span both `pbr-tools.test.js` and `integration.test.js`.
-
-When adding a new hook script, create a corresponding test file. Tests must pass on Windows, macOS, and Linux.
-
-## Dashboard
-
-The dashboard (`dashboard/`) is a separate Express.js application with its own dependency tree (`dashboard/package.json`). It provides a web UI for browsing `.planning/` state. Tech: Express 5.x, EJS, Pico.css, HTMX 2.0, chokidar for file watching, SSE for live updates.
-
-Dashboard tests use Vitest (not Jest) and live in `dashboard/tests/`. Run with `npm --prefix dashboard test`.
+# CLAUDE.md
+
+This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
+
+## What This Is
+
+Plan-Build-Run is a **Claude Code plugin** that provides a structured development workflow. It solves context rot — quality degradation as Claude's context window fills up — through disciplined subagent delegation, file-based state, and goal-backward verification. Users invoke `/pbr:*` slash commands (skills) that orchestrate specialized agents via `Task()`.
+
+## Critical Rules
+
+- **NEVER add AI co-author lines** to git commits or PRs. No `Co-Authored-By: Claude` or similar. Only add co-author lines referencing actual human contributors.
+- **NEVER inline agent definitions** into skill prompts. Use `subagent_type: "pbr:{name}"` — Claude Code auto-loads agent definitions from `agents/`. Reading agent `.md` files wastes main context.
+
+## Commands
+
+```bash
+npm test                                    # Run all Jest tests (~780 tests, 36 suites)
+npm run lint                                # ESLint on plugins/pbr/scripts/ and tests/
+npm run validate                            # Validate plugin directory structure
+npx jest tests/validate-commit.test.js      # Run a single test file
+npx jest --coverage                         # Run with coverage report
+```
+
+Coverage thresholds (enforced in `package.json`): 65% statements, 58% branches, 70% functions, 65% lines.
+
+Dashboard (separate dependency tree):
+```bash
+npm run dashboard:install                   # One-time install of dashboard deps
+npm run dashboard -- --dir /path/to/project # Launch dashboard for a project
+```
+
+Load the plugin locally for manual testing:
+```bash
+claude --plugin-dir .
+```
+
+CI runs on Node 18/20/22 across Windows, macOS, and Linux. All three platforms must pass.
+
+## Architecture
+
+All plugin code lives under `plugins/pbr/`. Three layers:
+
+### Skills (`skills/{name}/SKILL.md`)
+Markdown files with YAML frontmatter defining slash commands (`/pbr:begin`, `/pbr:plan`, etc.). Each SKILL.md is a complete prompt that tells the orchestrator what to do. Skills read state, interact with the user, and spawn agents.
+
+21 skills: begin, build, config, continue, debug, discuss, explore, health, help, import, milestone, note, pause, plan, quick, resume, review, scan, setup, status, todo.
+
+### Agents (`agents/{name}.md`)
+Markdown files with YAML frontmatter defining specialized subagent prompts. Agents run in fresh `Task()` contexts with clean 200k token windows. Spawned via `subagent_type: "pbr:{name}"` — auto-loaded by Claude Code.
+
+10 agents: researcher, planner, plan-checker, executor, verifier, integration-checker, debugger, codebase-mapper, synthesizer, general.
+
+### Scripts (`scripts/*.js`)
+28 Node.js hook scripts that fire on Claude Code lifecycle events. Configured in `hooks/hooks.json`. All use CommonJS, must be cross-platform (`path.join()`, not hardcoded separators), and log via `logHook()` from `hook-logger.js`.
+
+**Dispatch pattern**: Several hooks use dispatch scripts that fan out to sub-scripts based on the file being written/read:
+
+| Hook Event | Entry Script | Delegates To |
+|------------|-------------|-------------|
+| SessionStart | progress-tracker.js | — (injects project state) |
+| PostToolUse (Write\|Edit) | post-write-dispatch.js | check-plan-format.js, check-roadmap-sync.js, check-state-sync.js |
+| PostToolUse (Write\|Edit) | post-write-quality.js | check-doc-sprawl.js, check-skill-workflow.js |
+| PostToolUse (Task) | check-subagent-output.js | — (validates agent output) |
+| PostToolUse (Write\|Edit) | suggest-compact.js | — (context budget warnings) |
+| PostToolUse (Read) | track-context-budget.js | — (tracks reads for budget) |
+| PostToolUseFailure | log-tool-failure.js | — (logs failures) |
+| PreToolUse (Bash) | pre-bash-dispatch.js | validate-commit.js, check-dangerous-commands.js, check-phase-boundary.js |
+| PreToolUse (Write\|Edit) | pre-write-dispatch.js | — (write guards) |
+| PreCompact | context-budget-check.js | — (preserves STATE.md) |
+| Stop | auto-continue.js | — (chains next command) |
+| SubagentStart/Stop | log-subagent.js | — (tracks lifecycle) |
+| SubagentStop | event-handler.js | — (auto-verification trigger) |
+| TaskCompleted | task-completed.js | — (processes task completion) |
+| SessionEnd | session-cleanup.js | — (cleanup) |
+
+**Hook exit codes**: 0 = success, 2 = block (PreToolUse hooks that reject a tool call).
+
+**`${CLAUDE_PLUGIN_ROOT}`**: Used in hooks.json to reference script paths. Claude Code expands this internally — works on all platforms without shell expansion.
+
+### Supporting directories
+
+- **`references/`** — Shared reference docs loaded by skills (plan format, commit conventions, UI formatting, deviation rules)
+- **`templates/`** — EJS-style `.tmpl` files for generated markdown (VERIFICATION.md, SUMMARY.md, etc.)
+- **`commands/`** — Command registration files (one `.md` per command mapping to its skill)
+- **`skills/shared/`** — 14 shared skill fragments extracted from repeated patterns across skills (config-loading, digest-select, revision-loop, context-loader-task, universal-anti-patterns, phase-argument-parsing, gate-prompts, state-loading, state-update, commit-planning-docs, context-budget, domain-probes, progress-display, error-reporting)
+
+## Key Conventions
+
+**Commit format**: `{type}({scope}): {description}` — enforced by PreToolUse hook. Types: feat, fix, refactor, test, docs, chore, wip. Scopes: `{NN}-{MM}` (phase-plan), `quick-{NNN}`, `planning`, `tools`, or descriptive word.
+
+**Skill frontmatter** (SKILL.md):
+```yaml
+---
+name: skill-name
+description: "What this skill does"
+allowed-tools: Read, Write, Edit, Bash, Glob, Grep, Task
+argument-hint: "<N> [--flag]"
+---
+```
+
+**Agent frontmatter** ({name}.md):
+```yaml
+---
+name: agent-name
+description: "What this agent does"
+model: sonnet|inherit|haiku
+memory: none|user|project
+tools:
+  - Read
+  - Write
+  - Bash
+---
+```
+
+## Data Flow
+
+Skills and agents communicate through files on disk, not messages:
+
+```
+.planning/STATE.md      ← source of truth for current position
+.planning/ROADMAP.md    ← phase structure, goals, dependencies
+.planning/config.json   ← workflow settings (~62 properties across 12 top-level keys)
+.planning/phases/NN-slug/
+  PLAN.md               ← written by planner, read by executor
+  SUMMARY.md            ← written by executor, read by orchestrator
+  VERIFICATION.md       ← written by verifier, read by review skill
+```
+
+The orchestrator stays lean (~15% context) by delegating heavy work to agents. Each agent gets a fresh context window.
+
+**Utility library**: `pbr-tools.js` is a shared Node.js library (stateLoad, configLoad, frontmatterParse, mustHavesCollect, etc.) used by multiple hook scripts. It provides CLI subcommands that agents call to avoid wasting tokens on file parsing.
+
+## Testing
+
+Tests live in `tests/` using Jest. Test files mirror script names: `validate-commit.test.js` tests `validate-commit.js`.
+
+**Fixture project**: `tests/fixtures/fake-project/.planning/` provides read-only fixture data for tests.
+
+**Mutation tests**: Use `fs.mkdtempSync()` to create temporary directories — never mutate the fixture project.
+
+**`pbr-tools.js` tests** span both `pbr-tools.test.js` and `integration.test.js`.
+
+When adding a new hook script, create a corresponding test file. Tests must pass on Windows, macOS, and Linux.
+
+## Dashboard
+
+The dashboard (`dashboard/`) is a separate Express.js application with its own dependency tree (`dashboard/package.json`). It provides a web UI for browsing `.planning/` state. Tech: Express 5.x, EJS, Pico.css, HTMX 2.0, chokidar for file watching, SSE for live updates.
+
+Dashboard tests use Vitest (not Jest) and live in `dashboard/tests/`. Run with `npm --prefix dashboard test`.

package/LICENSE

@@ -1,21 +1,21 @@
-MIT License
-
-Copyright (c) 2025 Towline Contributors
-
-Permission is hereby granted, free of charge, to any person obtaining a copy
-of this software and associated documentation files (the "Software"), to deal
-in the Software without restriction, including without limitation the rights
-to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
-copies of the Software, and to permit persons to whom the Software is
-furnished to do so, subject to the following conditions:
-
-The above copyright notice and this permission notice shall be included in all
-copies or substantial portions of the Software.
-
-THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-SOFTWARE.
+MIT License
+
+Copyright (c) 2025 Towline Contributors
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.