18 tools with significant quality gaps. Most tools have basic descriptions (50-150 chars) but lack parameter-level documentation. Input schemas are present but incomplete: many parameters lack type definitions or descriptions. No output schemas documented. Error handling is absent, tools do not guide recovery or categorize failures. Tool naming is inconsistent (ai_search vs brave_llm_context vs context_brand_intel). Composition issues: tools like search_and_read and context_web_extract combine multiple concerns. No evidence of idempotency, pagination guidance, or field-level chaining IDs in responses. Security: API keys injected via environment (correct), but no audit logging or permission gates visible.

• Add output schema documentation for all 18 tools. Specify field names, types, and which fields are required for chaining (e.g., result_id, url, snippet). Use JSON Schema format in tool metadata.
• Add parameter-level descriptions for all parameters. For enums (provider, action, context_type, search_type), list valid values explicitly. For numeric params (limit, max_depth), specify min/max ranges.
• Split composite tools: separate search_and_read into search_web + read_url; separate context_web_extract into extract_content + optimize_for_llm. Each tool should do one thing.
• Implement error handling with recovery guidance. Return structured errors: {code, message, retryable, nextSteps}. E.g., 'Provider rate-limited. Retry in 60s or try provider=brave instead.'
• Standardize naming: rename context_* tools to get_* or extract_* (e.g., extract_brand_intel, extract_styleguide). Rename result_read to retrieve_result. All tools should start with action verbs.