earakely-scale/MetabaseEnv-REA

metabase

metabase

An AgentEnv bundle with its recorded runs, published with agentenv-hf.

Tables

• metabase_tasks: one row per task of the bundle, with its prompt, envs, verifiers and steps.
• metabase_episodes: one row per recorded run, with its status, model, reward and scores, the prompt and response, and the agent's transcript as chat messages when its trajectory format is one agentenv-hf reads.

raw/metabase.jsonl holds each run's record and native trajectory, and bundles/metabase/ the bundle itself.

Run it

pip install agentenv-framework huggingface_hub
hf download earakely-scale/MetabaseEnv-REA --repo-type dataset --local-dir agentenv-dataset
agent-env run agentenv-dataset/bundles/metabase

Metabase on AgentEnv, from a REA capture

What this is. A replay environment of Metabase 0.64.2.8 (AGPL-3.0), built from one recorded browser session against a self-hosted instance seeded with synthetic data. REA projected the capture into an environment manifest (42 tools), and rea environment-export wrote the bundle you can run from this dataset. Every tool call replays a retained record from the capture, cites the record and the Evidence it came from, and nothing executes Metabase. Tasks are graded deterministically, with no judge.

What's in it

• 42 tools (37 GET, 3 POST, 1 DELETE, 1 PUT) over 120 recorded exchanges on http://127.0.0.1:3201; no recorded group excluded.
• Path parameters: 25 numeric. REA's four format rules (numeric, uuid, hex, iso-date) applied unconditionally; the three opt-in rules (after-uid-literal, opaque-token, sibling-variance) were requested and are labelled per parameter. With the four format rules alone the manifest has 42 tools.
• Unknowns the manifest states explicitly: authorization model, response semantics beyond recorded exchanges, state transitions between tools.
• 12 retrieval tasks (evals/metabase-retrieval.toml): each asks one factual question about a seeded object, with a distinct answer that the app does not ship by default, answerable through a recorded exchange; reward 1 if the response contains the recorded value, else 0. Plus smoke, which deploys the environment and checks replay through the MCP endpoint.
• 24 recorded episodes by openai/gpt-oss-120b:groq, zai-org/GLM-5.3-Flash:zai-org through the Hugging Face router, with their chat transcripts.
• Provenance: capture ev_ae41d6f2a19b032ad77cd55df2547050605d49d7877db2e63b47c49647324924 (scoped HAR sha256 cd791130c103bf19299b14221ac2c27ae0bc9f8ae16e604c82ac4b34631a2d8a, 515,053 bytes), manifest view ev_97964117e7834ff3666205d78ca1a191235de72b54f4aa6bc0f96899674581a8, confidence inferred, authority historical-reference. The REA build is a prerelease of the environment commands proposed upstream in morluto/rea#4954, #4955, #4975 and #4999; the bundle installs it from the tarball it ships, so the environment runs without waiting for an npm release.

What replay means, and does not

A tool call returns the recorded response whose path and query values best match the arguments, with replay.exact_match telling you whether it matched exactly or fell back to the closest record. Reads of seeded objects behave like the application. Writes return the recorded answer of the one write the session made, and the environment has no state: a created object does not appear in a later list. This is what makes retrieval tasks hard-gradable and write tasks weak; the manifest's unknowns say so.

Tasks and grading

task	tool	arguments	field	answer
qa-01-get_permissions_group_by_id	get_permissions_group_by_id	id=5	members[0].email	ada.finch@example.test
qa-02-get_timeline_by_id	get_timeline_by_id	id=1, include=events	events[0].name	Quarry audit signed off by Rui Castellanos
qa-03-get_dashboard_by_id	get_dashboard_by_id	id=2	description	Tide-table menu metrics for the 36-cover sittings.
qa-04-get_timeline_event_by_id	get_timeline_event_by_id	id=1	description	412 blocks north quarry, 97 in the tidal yard.
qa-05-get_card_by_id	get_card_by_id	id=2	description	Mean product rating by category, read out at the 22:00 kitchen handover.
qa-06-get_native_query_snippet_by_id	get_native_query_snippet_by_id	id=1	name	Saltmarsh Depot gizmo filter
qa-07-get_collection_items_by_id	get_collection_items_by_id	id=7	data[0].name	Blue Heron Kitchen average rating per category
qa-08-get_segment_by_id	get_segment_by_id	id=1	description	Orders whose ledger total is above 120 tide-tokens.
qa-09-get_dashboard	get_dashboard	f=all	[0].name	Blue Heron Kitchen Service Board
qa-10-get_segment	get_segment	no arguments	[0].name	Granite Ledger large orders
qa-11-get_timeline	get_timeline	include=events	[0].description	Audit dates and depot changes for the ledger.
qa-12-get_native_query_snippet	get_native_query_snippet	no arguments	[0].content	CATEGORY = 'Gizmo' AND RATING >= 4

Episodes

model	tasks	passed	pass rate	mean tool calls	mean input tokens
openai/gpt-oss-120b:groq	12	12	100%	1.0	10,358
zai-org/GLM-5.3-Flash:zai-org	12	12	100%	1.0	13,478

Every episode is in the episodes config with its transcript; rewards are the deterministic verifier scores.

Every recorded episode passed.

Run it yourself

uv tool install agentenv-framework --with 'agentenv-hf @ git+https://github.com/earakely-scale/agentenv-hf-plugin@v0.3.0'
agent-env hf run earakely-scale/MetabaseEnv-REA --bundle metabase --task smoke          # Docker, no model
agent-env hf run earakely-scale/MetabaseEnv-REA --bundle metabase --eval metabase-retrieval --model openai/gpt-oss-120b:groq

The retrieval eval needs a model endpoint: set [model] base_url and api_key in .agentenv/config.toml (the Hugging Face router works with any hf_ token). The bundle's agents/replay-qa is a small OpenAI-compatible tool-calling agent; any A2A agent can replace it.

How it was made

1. Metabase ran in Docker on a throwaway VM, bound to loopback, seeded through its own API with synthetic records (site Lighthouse Rota Analytics with admin Rea Admin; users Ada Finch, Rui Castellanos and Mina Okafor; group Night Shift Analysts; collections Granite Ledger, Blue Heron Kitchen and Saltmarsh Depot; 5 saved questions on the bundled Sample Database (MBQL, native SQL and a model); dashboards Granite Ledger Overview and Blue Heron Kitchen Service Board; timeline Granite Ledger milestones with 2 events; native-query snippet Saltmarsh Depot gizmo filter; segment Granite Ledger large orders; synthetic descriptions on 1 table and 2 fields); people and projects are invented.
2. Playwright drove a login, the main pages and a scripted read, write, read sequence through the app's JSON API; the HAR was scoped to the API, the login exchange dropped, secret-looking body fields scrubbed, headers and cookies removed.
3. rea inspect-web-network-capture retained the exchanges as Evidence; rea inspect-analysis-view projected the environment-manifest view; rea environment-export wrote this bundle; gen_tasks.py derived the retrieval tasks from the manifest and the recorded bodies; agent-env run recorded the episodes; agent-env hf publish wrote this dataset. Scripts: https://github.com/earakely-scale/rea-environments.

Compared with PortSimEnv on AgentEnv

	PortSimEnv-AgentEnv	Metabase (this dataset)
Environment	A simulation: a container quay, ships, cranes, weather, with A2A ship agents	A replay of a real application's API, captured from one session
Observation	Tool results computed by the simulator as the week unfolds	Recorded responses, cited to Evidence
Action space	Planning tools (check_plan, confirm_berths, advance, message_agent)	42 tools mirroring the application's endpoints
State	Full: a virtual clock, triggers, frozen windows, agents that answer	None: every call is independent; writes replay one recorded answer
Reward	Deterministic, against a CP-SAT optimum or a re-planner anchor; continuous 0 to 1	Deterministic, recorded-value match; 0 or 1 per task
What it measures	Planning under uncertainty and negotiation	Tool discovery, parameter inference and faithful retrieval in a real API
Data provenance	Real 2024 port calls and weather; simulated workloads	Synthetic seed data on a self-hosted Metabase; nothing real
How it was built	Hand-written plugin, months of work	Generated from a capture in one scripted pass
Where it is strong	Long-horizon decisions with hard rewards	Breadth: any app you can capture becomes an environment

The two are complementary: PortSimEnv shows what a hand-built simulation can grade; this dataset shows how far a capture alone gets you, and the manifest's unknowns list exactly what a stateful version would still have to specify.

Licence and credits

Bundle code and tasks: Apache-2.0. Recorded content: responses generated by a self-hosted Metabase instance over synthetic data; Metabase is AGPL-3.0 software by Metabase, Inc. and is not redistributed here beyond the recorded JSON responses. REA is MIT software by morluto; AgentEnv Framework is Apache-2.0 by Scale AI.