{"id":"temporal-mcp-server","name":"temporal-mcp-server","af_score":51.2,"security_score":55.5,"reliability_score":26.2,"what_it_does":"Provides a Model Context Protocol (MCP) server exposing tools for managing Temporal workflow orchestration (start/query/control/history) and schedule management. Supports running against local or remote Temporal instances, with Docker or PyPI-based execution, and configurable connection/TLS/API key settings.","best_when":"When you want a standardized MCP interface to Temporal that can be consumed by MCP clients (e.g., IDEs/agents) to orchestrate workflows and schedules programmatically.","avoid_when":"When you require a fully documented REST/SDK experience, OAuth-style delegated authorization, or guaranteed idempotency/retry semantics for every operation.","last_evaluated":"2026-04-04T21:42:24.410203+00:00","has_mcp":true,"has_api":false,"auth_methods":["TEMPORAL_API_KEY / --api-key (API key)","mTLS client cert/key via TEMPORAL_TLS_CLIENT_CERT_PATH and TEMPORAL_TLS_CLIENT_KEY_PATH (client-side certificates)"],"has_free_tier":false,"known_gotchas":["Operations like start_workflow/control actions are likely not inherently idempotent; agents may need to avoid duplicate starts and handle retries carefully.","Batch operations rely on query filters; agents should ensure queries are precise to avoid unintended workflow selection.","Pagination is indicated via limit/skip for list endpoints; agents should loop until completion to avoid truncated results.","Tool semantics (e.g., behavior of cancel/terminate/continue-as-new, and how results are represented) are not described in README, so agents may need to validate against Temporal behavior."],"error_quality":0.0}