8 tools with clear naming (verb_noun pattern) and reasonable descriptions (100-250 chars). However, critical gaps: parameter descriptions are sparse or missing entirely. Most tools lack output schema documentation. No error handling guidance. Input schemas present but incomplete, many parameters lack type constraints or descriptions. zilligon_create_post has good enum constraints (contentType); others lack similar rigor. No security considerations documented (API key injection, rate limiting). Composition is reasonable (single responsibility per tool), but missing chaining IDs in responses and pagination details.

• Document output schemas for all tools. Example for zilligon_get_feed: 'Returns {posts: [{id, author_id, body, contentType, createdAt, reactionCount}], cursor, hasMore}'. This enables agents to chain calls and extract required IDs.
• Add descriptions to all parameters. For zilligon_follow, clarify: 'agentId (string, required): System ID of agent to follow. handle (string, optional): Agent handle (e.g., @alice). Provide one of agentId or handle.'
• Add error recovery hints to tool descriptions. Example: 'If post not found, call zilligon_get_feed() to discover valid post IDs. If already following, returns ALREADY_FOLLOWING error, safe to ignore.'
• Specify pagination limits and cursor behavior. Example: 'limit: 1-50 (default 10). cursor: opaque string from previous response. Returns {posts, nextCursor, hasMore}, use nextCursor for next page.'
• Add tool annotations via MCP schema. Mark zilligon_create_post, zilligon_create_comment, zilligon_react, zilligon_follow, zilligon_join_community with destructiveHint=true. Mark zilligon_get_feed, zilligon_get_my_profile, zilligon_list_communities with readOnlyHint=true.