22 tools with consistent verb_noun naming (kanka_list_*, kanka_get_*, kanka_create_*, kanka_update_*, kanka_delete_*). All tools have descriptions (10-80 chars, mostly adequate). Input schemas present for all tools with type definitions and parameter descriptions. However, descriptions are terse (avg ~40 chars vs baseline 194 chars), lack WHEN/WHY context, and omit dependency hints. Object parameters ('data' fields) lack nested schema documentation. Error handling guidance absent. No tool annotations (readOnlyHint, destructiveHint). Output schemas not documented in tool definitions.

• Expand tool descriptions to 100-200 chars. Include WHEN to call (e.g., 'Call after list_campaigns to fetch a specific campaign's details'), WHAT it returns (e.g., 'Returns campaign name, description, and member count'), and any prerequisites (e.g., 'Requires authentication via kanka_oauth_login or KANKA_TOKEN env var').
• Add tool annotations: mark kanka_delete_* with destructiveHint=true, kanka_list_* and kanka_get_* with readOnlyHint=true, and idempotent tools with idempotentHint=true. This enables LLMs to reason about side effects.
• Document output schemas for all tools. For list_campaigns, specify: returns array of {id, name, description, member_count}. For get_entity, specify: returns {id, type, name, entry, ...} with all top-level fields typed.
• For create/update tools, replace generic 'data' parameter with explicit nested schemas. E.g., kanka_create_entity should accept entity_type enum + typed fields (character: {name, title, age}, location: {name, description}, etc.) or link to kanka_describe_entity_type for schema discovery.
• Add error recovery hints to descriptions. E.g., 'If campaign_id is invalid, call kanka_list_campaigns to discover available campaigns.' This guides LLMs to self-correct without dead-ending.

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

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