SSWP MCP has well-structured schemas with proper JSON Schema validation (AJV), clear risk classifications (WRITE/READ_ONLY), and explicit approval-gating for sensitive operations. However, parameter descriptions are sparse or missing entirely, most params lack explanatory text beyond type constraints. Tool descriptions are domain-specific but assume deep familiarity with SSWP attestation concepts; they would benefit from clearer WHEN/WHY guidance for LLM selection. No output schemas are documented, forcing LLMs to infer response structure. Error handling is present but minimal, no recovery guidance or actionable error messages visible in the dispatch logic.

• Add parameter descriptions to every input field. For 'repoPath', explain: 'Absolute filesystem path to the repository root (must be canonical, no symlinks).' For 'approval', document the expected structure: 'One-use operator-policy approval receipt containing MAC and task binding.'
• Document output schemas for all tools. Example for sswp_witness: 'Returns {attestation: {id, target, checks, timestamp}, delivery: "queued_for_brain"|"pending", materialization: {status}}'. Include field types and meanings.
• Enhance error messages with recovery guidance. Instead of throwing 'Invalid input', return structured errors: {error: 'Invalid approval', reason: 'MAC verification failed', nextStep: 'Request a new approval from omega_authorize_action'}.
• Simplify tool descriptions for LLM clarity. Add a 'WHEN' clause: 'Use sswp_witness to run approved build checks on a repository. Call this after obtaining an approval receipt from omega_authorize_action.' Reduce jargon or define it inline.
• Add enum constraints or pattern validation for 'query' parameter in sswp_node_search. Document whether it matches literal names, descriptions, or paths, and whether matching is case-sensitive or fuzzy.