Initial commit: OpenClaw Skill Collection

6 custom skills (assign-task, dispatch-webhook, daily-briefing,
task-capture, qmd-brain, tts-voice) with technical documentation.
Compatible with Claude Code, OpenClaw, Codex CLI, and OpenCode.

openclaw-knowhow-skill/docs/infrastructure/hooks/index.md

@@ -0,0 +1,41 @@
+# Hooks Documentation Summary
+
+## Core Concept
+OpenClaw's hooks system provides an event-driven mechanism for automating actions in response to agent commands and lifecycle events. Hooks are small scripts that run when something happens.
+
+## Key Components
+
+**Two Hook Categories:**
+- **Hooks**: Run inside the Gateway when agent events fire (like `/new`, `/reset`, `/stop`)
+- **Webhooks**: External HTTP endpoints for triggering work in OpenClaw
+
+**Discovery Locations (by precedence):**
+1. Workspace hooks (`<workspace>/hooks/`)
+2. Managed hooks (`~/.openclaw/hooks/`)
+3. Bundled hooks (shipped with OpenClaw)
+
+## Structure Requirements
+
+Each hook requires:
+- **HOOK.md**: Metadata in YAML frontmatter plus documentation
+- **handler.ts**: TypeScript implementation exporting a `HookHandler` function
+
+The metadata object supports fields like `emoji`, `events`, `requires` (for binaries/environment variables), and `homepage`.
+
+## Event Types
+
+**Command events**: `command:new`, `command:reset`, `command:stop`
+**Agent events**: `agent:bootstrap`
+**Gateway events**: `gateway:startup`
+
+## Bundled Hooks
+
+Four hooks ship with OpenClaw:
+- **session-memory**: Saves context when `/new` is issued
+- **command-logger**: Audit trail to `~/.openclaw/logs/commands.log`
+- **boot-md**: Executes `BOOT.md` on gateway startup
+- **soul-evil**: Swaps SOUL content during specified windows
+
+## Management
+
+CLI commands include `openclaw hooks list`, `enable`, `disable`, `info`, and `check` for discovery and eligibility verification.

openclaw-knowhow-skill/docs/infrastructure/hooks/soul-evil.md

@@ -0,0 +1,30 @@
+# SOUL Evil Hook Documentation
+
+## Overview
+
+The SOUL Evil hook is a feature that replaces injected `SOUL.md` content with `SOUL_EVIL.md` during specified conditions, operating entirely in memory without modifying disk files.
+
+## Core Functionality
+
+When `agent:bootstrap` runs, the hook can replace the `SOUL.md` content in memory before the system prompt is assembled. If the evil variant is unavailable, the system logs a warning and maintains the standard SOUL configuration.
+
+An important limitation: this hook does not function for sub-agent runs, as they exclude `SOUL.md` from their bootstrap files.
+
+## Configuration
+
+Enable the hook via command:
+```bash
+openclaw hooks enable soul-evil
+```
+
+Then configure using JSON with these parameters:
+- `file`: alternate SOUL filename (defaults to `SOUL_EVIL.md`)
+- `chance`: probability value between 0–1 for random activation
+- `purge.at`: scheduled daily trigger in HH:mm format (24-hour)
+- `purge.duration`: window length (examples: `30s`, `10m`, `1h`)
+
+The purge window takes precedence over random chance when both are active.
+
+## Key Characteristics
+
+The hook respects the timezone setting from `agents.defaults.userTimezone`, falling back to host timezone if unspecified. It requires creating `SOUL_EVIL.md` in the agent workspace root directory alongside the standard SOUL file.