← ClaudeAtlas

runtype-adminlisted

Use when operating a live Runtype account through MCP or Code Mode MCP: inspect resources, debug failed flows or agents, read logs and traces, manage products, surfaces, records, schedules, secrets, models, evals, batches, conversations, client tokens, or make safe account mutations. Includes search/execute patterns, read-before-write, validation, pagination, and conservative destructive-change policy.
runtypelabs/skills · ★ 1 · AI & Automation · score 68
Install: claude install-skill runtypelabs/skills
# Runtype Admin Use this skill for live account operations. All mutations affect the user's Runtype workspace unless explicitly run against a disposable test account. ## Operating Stance - Reads can proceed after discovering the target. - Single-resource writes the user explicitly requested can proceed after reading current state and validating payloads. - Multi-resource builds should use `runtype-build-product`. - Destructive, bulk, or hard-to-reverse actions require explicit confirmation. - Never claim a mutation succeeded unless the tool result confirms it. - Never show secret values. Use metadata only. ## Standard MCP For direct MCP tools, discover before acting: - Identity and inventory: `get_me`, `list_products`, `list_agents`, `list_flows`, `list_tools`, `list_records`, `list_schedules`, `list_secrets`, `list_model_configs`, `list_conversations`. - Debugging: `list_logs`, `get_log_stats`, `trace_execution`, `trace_conversation`, `list_agent_executions`, `get_record_results`, `get_record_step_results`, `get_record_costs`. - Validation: `validate_flow`, `validate_code`, `validate_product`, `validate_product_flow`, `validate_product_agent`, `validate_product_surface`, `validate_product_tool`. - Evals and batches: `submit_eval`, `list_eval_batches`, `get_eval_results`, `compare_eval`, `compare_eval_record`, `analyze_eval_steps`, `submit_batch`, `get_batch_status`, `get_batch_summary`, `cancel_batch`. Surfaces are product-scoped. List or mutate them