Reddit Ads MCP has basic tool structure but significant quality gaps. All 8 tools have names and descriptions, but descriptions are minimal (10-50 chars, well below the 194-char baseline). Input schemas are present with type definitions, but parameter descriptions are sparse or missing entirely. No output schemas documented. Error handling is minimal, most tools throw generic errors without recovery guidance. The server lacks pagination support despite tools returning lists (campaigns, ad groups, posts). No tool annotations (readOnlyHint/destructiveHint) despite clear risk distinctions. create_campaign is the only write operation but has no confirmation or dry-run pattern.

• Expand all tool descriptions to 100-200 chars. Include WHAT the tool does, WHEN to use it (vs similar tools), and what it returns. Example: 'Retrieve all campaigns for an ad account. Use this to list active campaigns before creating ad groups. Returns campaign ID, name, status, objective, and funding instrument ID.'
• Document output schemas for all tools. Define the structure of AdAccount, Campaign, AdGroup, Profile, Post, FundingInstrument as JSON Schema in tool definitions. Include all fields the LLM might need for chaining (e.g., campaign_id in get_campaigns response so agent can pass it to get_ad_groups).
• Add pagination to list tools: accept limit (1-100, default 20) and offset/cursor parameters. Return total_count and next_cursor. Cap results at 50 items per call to prevent context window exhaustion.
• Add parameter descriptions for all inputs. Example: 'adAccountId (string, required): The Reddit Ads account ID. Obtain from get_ad_accounts().' Describe constraints: 'spendCap (number, optional): Daily spend limit in USD. Must be >= 1.00.'
• Add tool annotations: mark get_* tools with readOnlyHint=true; mark create_campaign with destructiveHint=true. Add idempotentHint=true to read-only tools so agents know they can retry safely.