18 tools with consistent naming (verb_noun pattern: list_, get_, search_) and reasonable descriptions (avg ~80 chars). All tools have input schemas with typed parameters and descriptions. However, output schemas are not documented in the source code, responses are inferred from API behavior, not explicitly declared. Error handling is minimal; no recovery guidance or actionable error messages visible. Missing tool annotations (readOnlyHint, destructiveHint). One write tool (lightspeed_set_config) lacks confirmation/dry-run pattern. Parameter descriptions are adequate but lack format constraints (e.g., date_from/date_to lack ISO 8601 format examples). No pagination metadata visible in list tools.

• Document output schemas for all tools. For each tool, add a 'Returns' section in the description or a separate schema field listing response object structure, field types, and required fields. Example: 'Returns: {sales: [{id, amount, date, outlet_id, ...}], total_count: number}'.
• Add tool annotations to the MCP schema. Wrap lightspeed_set_config with destructiveHint=true and all read-only tools with readOnlyHint=true. This enables clients to render safety warnings and enforce least-privilege policies.
• Implement a dry-run or preview mode for lightspeed_set_config. Add an optional 'preview' parameter (boolean, default false). When true, return the proposed changes without applying them. This prevents accidental configuration overwrites.
• Add format constraints to date parameters. Use regex pattern '^\d{4}-\d{2}-\d{2}$' or an enum of relative date strings (today, yesterday, last_7_days, last_30_days) to guide LLM input and reduce validation errors.
• Document pagination in list tools. Add 'offset' and 'total_count' to responses, or use cursor-based pagination. Update tool descriptions to state: 'Returns up to [limit] results. Use offset to fetch additional pages. total_count indicates total available records.'