Single tool 'systemprompt' with critical definition gaps. Description is 287 chars (within baseline 194 avg), but lacks clarity on what the tool actually does beyond 'Execute SystemPrompt CLI commands.' Input schema present but minimal: only 'command' string param with generic description. No output schema documented. No error handling guidance. Tool name 'systemprompt' is a noun, not a verb, violates verb_noun naming convention. Description includes example values ('core playbooks show guide_start') which LLMs may reuse literally. No parameter constraints, ranges, or validation rules. No indication of idempotency, side effects, or when to use this vs alternatives. Composition is poor: single catch-all tool that wraps an entire CLI, forcing agents to understand SystemPrompt command syntax rather than abstracting it into domain-specific tools.

• Rename tool from 'systemprompt' to action-verb names: 'list_playbooks', 'show_playbook', 'send_discord_message', 'list_agents'. Each tool should wrap one CLI command, not the entire CLI.
• Document output schema for each tool. E.g., list_playbooks returns {playbooks: [{id, name, description}], total_count}. Baseline: 194 chars avg description, 100% of A+ tools have return types.
• Remove example values from descriptions. Replace 'e.g. core playbooks show guide_start' with enum constraints or format rules in the parameter schema.
• Add parameter constraints: if 'command' remains, define allowed command prefixes (core, plugins, admin) as an enum, or split into separate tools with specific parameters (playbook_id, channel_id, message_text).
• Add error handling guidance: 'If command fails, check syntax against SystemPrompt CLI docs. Common errors: missing quotes, invalid playbook ID, permission denied. Retry with corrected command or call list_playbooks to discover valid IDs.'

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry