HFAO — Hugging Face Agent Observatory

The named artifact (test, doc, code path) exists in the repo today and is runnable. Per-claim, not per-project — a claim can be shipped while the parent project is still in development.

• Storage plane (ClickHouse + DuckDB) with parity tests

  tests/acceptance/test_ac_6_storage_*.py

• Ingest plane (OTel GenAI + OpenInference normalizer, OTLP server, PII redaction, bounded buffer, body offload at 64 KiB)

  tests/acceptance/test_ac_5_wire.py, test_ac_7_ingest.py

• SPEC.md v1.0.0 locked; every commit cites a spec section

  SPEC.md + commit message convention

• MCP-native queryability across observability primitives (FastMCP Streamable HTTP, §9.2 read surface + gated score_observation write)

  packages/hfao/mcp_server/ + tests/acceptance/test_ac_9_mcp.py (v0.5.0, SPEC §9)

• Closed eval-trace loop (traces ↔ datasets ↔ scores ↔ monitors), with causal attribution, cost rollups, and retention

  packages/hfao/compute/ + tests/acceptance/test_ac_8_causal.py, test_ac_8_eval.py, test_ac_8_cost_monitor.py (v0.5.0, SPEC §8)

• Stage 2 counterfactual replay — driver-pluggable, Tier-1 frameworks (LangGraph / OpenAI Agents SDK / Claude Agent SDK)

  packages/hfao/compute/causal/counterfactual.py + tests/acceptance/test_ac_8_counterfactual.py (post-v0.5.0 main, SPEC §16 Q-20)

• Proactive anomaly surfacing (Insight schema + AnomalyEngine) and rule-based insight routing (subscriptions)

  packages/hfao/compute/anomaly.py, schema/insights.py, compute/routing.py + tests/acceptance/test_ac_8_insights.py, test_ac_8_routing.py (post-v0.5.0 main, SPEC §16 Q-18/Q-19)