Claims (25)
claim.01conf: unstatedtier: primary
§ 2. Recurring feature categories
1. Run/trace timeline & span hierarchy. Langfuse (https://langfuse.com/), LangSmith (https://docs.langchain.com/langsmith/observability), AgentOps (https://adk.dev/integrations/agentops), Arize Phoenix (https://arize.com/docs/phoenix). Per-run waterfall/span tree; drill from session → run → LLM call → tool. Universal. 2. Token & cost dashboards. Langfuse, AgentOps, W&B Weave, Helicone, builderz MC ("token use and cost views"), tenacitOS ("Cost Tracking"), MeisnerDan (spend limits + circuit breaker). Universal. 3. Eval / quality scores. Langfuse, LangSmith, Phoenix (datasets & experiments), Weave (leaderboards), builderz MC (evals). Score runs, compare experiments, regression-track. 4. Logs / error / alerting views. Langfuse (alerts), LangSmith (threshold alerting), AgentOps (loop detection via max-iterations failure), builderz MC (activity stream, alerts, logs), tenacitOS (Activity Feed). 5. Topology / graph of agents. AutoGen Studio (control-transition graph), Arize (dedicated Agent Graph tab), LaunchDarkly (node/edge graph with on-node metrics), builderz MC (relationship graph), tenacitOS (Office 3D with a desk per agent). 6. Replay / time-travel. LangGraph Studio (checkpoints, fork/resume), AgentOps (session replay), Langfuse (replay of production traces). — We already do a light "replay wave" on load. 7. Per-agent metrics / utilization. AgentOps (multi-/cross-session metrics), tenacitOS (per-agent activity heatmap), MeisnerDan (agent utilization, velocity, burndown), Langfuse (custom dashboards). 8. Task / work queue management. builderz MC (task pipeline, review, receipts), MeisnerDan (Eisenhower matrix, kanban, decisions queue), AutoGen Studio (Team Builder). — We have no first-class task objects; this is the biggest data-model gap. 9. Session drill-down. Every trace tool: click an entity → slide-over/detail panel. 10. Scheduling / cron management. tenacitOS (Cron Manager with weekly timeline + manual trigger). Relevant to our heartbeat/cron usage. 11. Governance & approvals. builderz MC (roles, API keys, approvals, audits, trust signals); MeisnerDan (approval workflows, spend limits). Lower priority for a read-only view.
claim.02conf: unstatedsingle sourcetier: secondary
§ 3. What "agent topology / org visualization" projects specifically do
Node-edge graph as a first-class view. LaunchDarkly: each node is an agent, each edge is an output handoff; edges "define how output passes from one agent to the next" (https://launchdarkly.com/blog/agent-graphs-multi-agent-ai-workflows).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.03conf: unstatedsingle sourcetier: secondary
§ 3. What "agent topology / org visualization" projects specifically do
A dedicated agent-graph surface. Arize AX added a "dedicated Agent Graph tab … to visualize and explore agent interactions within traces" (https://arize.com/blog/new-in-arize-ax-experiment-comparisons-better-data-visualization-and-a-dedicated-agent-graph-tab).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.04conf: unstatedsingle sourcetier: secondary
§ 3. What "agent topology / org visualization" projects specifically do
Message-flow / control-transition graph in the live playground. AutoGen Studio renders "visual representation of message flow through a control transition graph" while streaming (https://microsoft.github.io/autogen/stable/user-guide/autogenstudio-user-guide/index.html).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.05conf: unstatedsingle sourcetier: primary
§ 3. What "agent topology / org visualization" projects specifically do
Spatial org metaphor is proven. tenacitOS ships an "Office 3D — interactive 3D office with one voxel avatar per agent" (https://github.com/carlosazaustre/tenacitOS) — i.e. the exact "agents at desks" metaphor our MC uses, in 3D, alongside a normal tabbed dashboard.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.06conf: unstatedsingle sourcetier: primary
§ 3. What "agent topology / org visualization" projects specifically do
Relationship graph of knowledge/memory (not just agents): builderz MC "memory browser, relationship graph, skills registry" (https://github.com/builderz-labs/mission-control).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.07conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://langfuse.com/ — Langfuse features.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.08conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://docs.langchain.com/langsmith/observability — LangSmith observability (redirect target of docs.smith.langchain.com).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.09conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://github.com/agentops-ai/agentops — AgentOps README (replays, graphs, cost).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.10conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://adk.dev/integrations/agentops — AgentOps + ADK (span hierarchy, replay analytics) [search-sourced].
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.11conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://microsoft.github.io/autogen/stable/user-guide/autogenstudio-user-guide/index.html — AutoGen Studio interfaces.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.12conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://docs.crewai.com/en/introduction — CrewAI Crews vs Flows.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.13conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://docs.openhands.dev/ — OpenHands Agent Canvas (redirect target of docs.all-hands.dev).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.14conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://arize.com/docs/phoenix — Arize Phoenix (redirect target of docs.arize.com/phoenix).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.15conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://arize.com/blog/new-in-arize-ax-experiment-comparisons-better-data-visualization-and-a-dedicated-agent-graph-tab — dedicated Agent Graph tab.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.16conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://docs.helicone.ai/getting-started/quick-start — Helicone gateway/logging (redirect target of docs.helicone.ai).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.17conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://wandb.ai/site/articles/llm-observability — W&B LLM observability / Weave overview.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.18conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://qaskills.sh/blog/weave-llm-evaluation-tracing-guide-2026 — W&B Weave guide [secondary].
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.19conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://www.langchain.com/blog/langgraph-studio-the-first-agent-ide — LangGraph Studio agent IDE.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.20conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://docs.langchain.com/oss/python/langchain/frontend/time-travel — LangGraph time-travel/checkpoints [search-sourced].
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.21conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://launchdarkly.com/blog/agent-graphs-multi-agent-ai-workflows — agent graphs + on-node metrics.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.22conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://github.com/builderz-labs/mission-control — self-hosted agent control plane, feature areas.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.23conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://github.com/MeisnerDan/mission-control — agent task command center, safety dashboard.
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.24conf: unstatedsingle sourcetier: primary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://github.com/carlosazaustre/tenacitOS — OpenClaw Mission Control dashboard (tabs + Office 3D).
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.
claim.25conf: unstatedsingle sourcetier: secondary
§ Appendix — sources (all fetched 2026-10-12 unless noted)
https://www.cnbc.com/2025/10/28/github-openai-google-anthropic-agents-ai.html — GitHub Agent HQ "mission control" [secondary press].
!
Single source — not cross-checked. Only 1 distinct source; the two-source rule is not met.