All 7 tools have basic descriptions (10-150 chars) and simple input schemas with type definitions. However, descriptions are minimal and lack actionable context for LLM selection. Parameter descriptions are present but generic. No output schemas documented. No error handling guidance. Tool names follow verb_noun pattern (deadline_search_*, deadline_get_*, deadline_list_*) but lack specificity about what data is returned or when to use each tool vs. alternatives. No enum constraints, no pagination guidance, no chaining IDs documented in responses.

• Document output schemas for all tools. Specify which fields are returned (e.g., job_id, step_id, task_id, session_id, logs) so LLMs can chain calls. Include total_count or next_cursor for pagination.
• Expand descriptions to 100-150 chars. Add WHEN to use each tool: 'Call this first to list all jobs, then deadline_get_job for details' or 'Use this to find failed steps after deadline_get_job shows failures.'
• Add enum constraints to task_run_status parameter in deadline_search_jobs. Document valid values (e.g., FAILED, SUCCEEDED, RUNNING) instead of free-form strings.
• Clarify the fallback pattern: rename deadline_get_session_logs to deadline_get_session_logs_fallback and document in both tools when to use each (primary vs. when worker logs unavailable).
• Add pagination parameters (limit, offset) and document result limits. E.g., 'Returns up to 50 jobs; use offset for pagination.' Include total_count in response schema.