Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="utf-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1"> | |
| <meta name="description" content="A recorded, synthetic trade-reconciliation example built with AgentEnv. Watch the walkthrough, inspect two recorded outcomes, and download the environment."> | |
| <title>TradeOpsSim — recorded AgentEnv walkthrough</title> | |
| <link rel="stylesheet" href="styles.css"> | |
| <script src="app.js" defer></script> | |
| </head> | |
| <body> | |
| <header class="masthead wrap"> | |
| <a class="brand" href="#top">TradeOpsSim <span>Built with AgentEnv</span></a> | |
| <a href="https://huggingface.co/datasets/AlexSchwartzman/TradeOpsSim-AgentEnv" target="_blank" rel="noopener">Environment & data ↗</a> | |
| </header> | |
| <main id="top" class="wrap"> | |
| <section class="intro" aria-labelledby="page-title"> | |
| <p class="eyebrow">Recorded demo · synthetic data</p> | |
| <h1 id="page-title">Test a reconciliation agent.</h1> | |
| <p class="lead">Your team defines the records, permitted tools and checks for a correct result. AgentEnv runs agents in that test environment and records the outcomes.</p> | |
| <video controls playsinline preload="metadata" poster="media/poster.png" aria-label="AgentEnv trade-reconciliation walkthrough"> | |
| <source src="media/tradeops-demo.mp4" type="video/mp4"> | |
| <track kind="captions" src="media/captions.vtt" srclang="en" label="English"> | |
| <a href="media/tradeops-demo.mp4">Download the video</a> | |
| </video> | |
| <div class="video-note"><span>2:32 walkthrough · Proposed GTS uses are labeled in the film.</span><a href="https://huggingface.co/datasets/AlexSchwartzman/TradeOpsSim-AgentEnv/blob/main/media/TRANSCRIPT.md" target="_blank" rel="noopener">Read transcript ↗</a></div> | |
| </section> | |
| <section class="recordings" aria-labelledby="cases-heading"> | |
| <p class="eyebrow">Inspect the recorded results</p> | |
| <h2 id="cases-heading">Two cases. Two correct outcomes.</h2> | |
| <p class="notice">This page replays existing results. <a href="https://huggingface.co/datasets/AlexSchwartzman/TradeOpsSim-AgentEnv" target="_blank" rel="noopener">Download the environment</a> to run your own agent.</p> | |
| <div class="tabs" role="tablist" aria-label="Recorded reconciliation cases"> | |
| <button id="tab-confirmed" type="button" role="tab" aria-selected="true" aria-controls="case-panel" data-case="0">Confirmed cancellation</button> | |
| <button id="tab-escalated" type="button" role="tab" aria-selected="false" aria-controls="case-panel" tabindex="-1" data-case="1">Missing confirmation</button> | |
| </div> | |
| <article id="case-panel" role="tabpanel" aria-labelledby="tab-confirmed" tabindex="0"></article> | |
| <noscript><p>The video and downloads work without JavaScript. Enable JavaScript to inspect the recorded case panels, or <a href="recorded-runs.json">download their evidence</a>.</p></noscript> | |
| <p class="scope">These are authored development cases, not a held-out benchmark. They do not establish production reliability or improvement over a rules baseline. No GTS systems or data were used.</p> | |
| </section> | |
| <section class="download" aria-labelledby="run-heading"> | |
| <p class="eyebrow">Run it yourself</p> | |
| <h2 id="run-heading">Keep the cases. Change the agent.</h2> | |
| <p>The dataset includes the runnable environment, thirteen authored cases, recorded evaluations and the application source. Start with the rules baseline below; it needs Python 3.11.4+, Git and Docker running.</p> | |
| <pre class="command"><code>python -m pip install 'agentenv-framework==0.9.1324' | |
| python -m pip install 'agentenv-hf @ git+https://github.com/earakely-scale/agentenv-hf-plugin@c116bf1b321fa33d671632b0ee1b77406dbe3b58' | |
| agent-env hf run AlexSchwartzman/TradeOpsSim-AgentEnv --bundle tradeops-baseline --task late_fill_duplicate</code></pre> | |
| <p>The baseline needs no model key. The <code>tradeops-agent</code> bundle supports a configured API model; follow its README for setup.</p> | |
| <a class="button" href="https://huggingface.co/datasets/AlexSchwartzman/TradeOpsSim-AgentEnv" target="_blank" rel="noopener">Open environment & instructions ↗</a> | |
| </section> | |
| </main> | |
| <footer class="wrap"><span>TradeOpsSim · Recorded runs · Synthetic data</span><a href="README.md">Sources & notices</a></footer> | |
| <script id="recorded-data" type="application/json">[ | |
| { | |
| "calls": [ | |
| { | |
| "arguments": {}, | |
| "name": "tradeops_read_runbook" | |
| }, | |
| { | |
| "arguments": {}, | |
| "name": "tradeops_get_workspace" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001" | |
| }, | |
| "name": "tradeops_recover_reports" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001", | |
| "report_id": "R-E2" | |
| }, | |
| "name": "tradeops_replay_execution" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001", | |
| "report_id": "R-CXL" | |
| }, | |
| "name": "tradeops_apply_order_status" | |
| } | |
| ], | |
| "capture": "captures/main-viewport.png", | |
| "capture_alt": "Recorded synthetic TradeOpsSim result: Recover the late fill. Book it once.", | |
| "checks": [ | |
| { | |
| "actual": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1" | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2" | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1" | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2" | |
| ] | |
| ], | |
| "id": "execution_membership", | |
| "label": "Every supported execution is booked once in its own scope", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1", | |
| 400 | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2", | |
| 200 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1", | |
| 400 | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2", | |
| 200 | |
| ] | |
| ], | |
| "id": "execution_quantities", | |
| "label": "Booked quantities match immutable source evidence", | |
| "passed": true | |
| }, | |
| { | |
| "actual": 2, | |
| "expected": 2, | |
| "id": "no_duplicate_bookings", | |
| "label": "A scoped execution never appears twice", | |
| "passed": true | |
| }, | |
| { | |
| "actual": { | |
| "cancel_request_status": "confirmed", | |
| "cum_qty": 600, | |
| "leaves_qty": 0, | |
| "status": "Cancelled" | |
| }, | |
| "expected": { | |
| "cancel_request_status": "confirmed", | |
| "cum_qty": 600, | |
| "leaves_qty": 0, | |
| "status": "Cancelled" | |
| }, | |
| "id": "order:ORD-1001", | |
| "label": "ORD-1001: cumulative fills, terminal status, and remaining quantity", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| "ORD-1001" | |
| ], | |
| "expected": [ | |
| "ORD-1001" | |
| ], | |
| "id": "order_membership", | |
| "label": "No unrelated order is removed or invented", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "ORD-1001", | |
| "ALPHA", | |
| "SIM-X", | |
| "2026-01-15", | |
| "ORD-1001", | |
| "SYNTH", | |
| "Buy", | |
| 1000 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "ORD-1001", | |
| "ALPHA", | |
| "SIM-X", | |
| "2026-01-15", | |
| "ORD-1001", | |
| "SYNTH", | |
| "Buy", | |
| 1000 | |
| ] | |
| ], | |
| "id": "order_scope_preserved", | |
| "label": "Original account, instrument, side, and order quantity are preserved", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "ALPHA", | |
| "SYNTH", | |
| 600 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "ALPHA", | |
| "SYNTH", | |
| 600 | |
| ] | |
| ], | |
| "id": "positions", | |
| "label": "Positions preserve account scope and buy/sell direction", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [], | |
| "expected": [], | |
| "id": "exception_disposition", | |
| "label": "Required unresolved issues remain open; resolved cases are not unnecessarily escalated", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "exception_evidence", | |
| "label": "Escalations cite observed evidence from the correct order", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "exception_source_check", | |
| "label": "The required source was queried before declaring its evidence incomplete", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "response_provenance", | |
| "label": "Cancellation dispositions are supported by observed authoritative responses", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [], | |
| "expected": [], | |
| "id": "safe_scope", | |
| "label": "No cross-scope or contradictory financial mutation was attempted", | |
| "passed": true | |
| }, | |
| { | |
| "actual": "1e8bf05b5076f445a22cba58815d4f746e2032a23177ea9673d9dea9afadab28", | |
| "expected": "1e8bf05b5076f445a22cba58815d4f746e2032a23177ea9673d9dea9afadab28", | |
| "id": "evidence_integrity", | |
| "label": "Authoritative source records remain immutable", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "evidence_provenance", | |
| "label": "Every booking refers to observed source evidence", | |
| "passed": true | |
| } | |
| ], | |
| "full_capture": "captures/main-full.png", | |
| "metrics": [ | |
| [ | |
| "600", | |
| "shares booked" | |
| ], | |
| [ | |
| "400", | |
| "shares cancelled" | |
| ], | |
| [ | |
| "0", | |
| "shares remaining" | |
| ], | |
| [ | |
| "14/14", | |
| "outcome checks passed" | |
| ] | |
| ], | |
| "model": "claude-sonnet-5-5", | |
| "run_id": "9216414cfa144869a6a9ce34b892ef9b", | |
| "scenario_id": "late_fill_duplicate", | |
| "summary": "Reconciled: every required outcome check passed.", | |
| "title": "Recover the late fill. Book it once." | |
| }, | |
| { | |
| "calls": [ | |
| { | |
| "arguments": {}, | |
| "name": "tradeops_read_runbook" | |
| }, | |
| { | |
| "arguments": {}, | |
| "name": "tradeops_get_workspace" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001" | |
| }, | |
| "name": "tradeops_inspect_order" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001" | |
| }, | |
| "name": "tradeops_recover_reports" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001", | |
| "report_id": "R-E2" | |
| }, | |
| "name": "tradeops_replay_execution" | |
| }, | |
| { | |
| "arguments": { | |
| "steps": 2 | |
| }, | |
| "name": "tradeops_wait_for_updates" | |
| }, | |
| { | |
| "arguments": { | |
| "order_key": "ORD-1001", | |
| "source": "order_status_archive" | |
| }, | |
| "name": "tradeops_recover_reports" | |
| }, | |
| { | |
| "arguments": { | |
| "code": "MISSING_CANCEL_ACK", | |
| "order_key": "ORD-1001", | |
| "report_ids": [ | |
| "R-E1", | |
| "R-E2", | |
| "R-E2-DUP" | |
| ] | |
| }, | |
| "name": "tradeops_open_exception" | |
| } | |
| ], | |
| "capture": "captures/escalated-viewport.png", | |
| "capture_alt": "Recorded synthetic TradeOpsSim result: Keep missing evidence unresolved.", | |
| "checks": [ | |
| { | |
| "actual": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1" | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2" | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1" | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2" | |
| ] | |
| ], | |
| "id": "execution_membership", | |
| "label": "Every supported execution is booked once in its own scope", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1", | |
| 400 | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2", | |
| 200 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E1", | |
| 400 | |
| ], | |
| [ | |
| "SIM-X", | |
| "2026-01-15", | |
| "ALPHA", | |
| "ORD-1001", | |
| "E2", | |
| 200 | |
| ] | |
| ], | |
| "id": "execution_quantities", | |
| "label": "Booked quantities match immutable source evidence", | |
| "passed": true | |
| }, | |
| { | |
| "actual": 2, | |
| "expected": 2, | |
| "id": "no_duplicate_bookings", | |
| "label": "A scoped execution never appears twice", | |
| "passed": true | |
| }, | |
| { | |
| "actual": { | |
| "cancel_request_status": "pending", | |
| "cum_qty": 600, | |
| "leaves_qty": 400, | |
| "status": "PartiallyFilled" | |
| }, | |
| "expected": { | |
| "cancel_request_status": "pending", | |
| "cum_qty": 600, | |
| "leaves_qty": 400, | |
| "status": "PartiallyFilled" | |
| }, | |
| "id": "order:ORD-1001", | |
| "label": "ORD-1001: cumulative fills, terminal status, and remaining quantity", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| "ORD-1001" | |
| ], | |
| "expected": [ | |
| "ORD-1001" | |
| ], | |
| "id": "order_membership", | |
| "label": "No unrelated order is removed or invented", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "ORD-1001", | |
| "ALPHA", | |
| "SIM-X", | |
| "2026-01-15", | |
| "ORD-1001", | |
| "SYNTH", | |
| "Buy", | |
| 1000 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "ORD-1001", | |
| "ALPHA", | |
| "SIM-X", | |
| "2026-01-15", | |
| "ORD-1001", | |
| "SYNTH", | |
| "Buy", | |
| 1000 | |
| ] | |
| ], | |
| "id": "order_scope_preserved", | |
| "label": "Original account, instrument, side, and order quantity are preserved", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "ALPHA", | |
| "SYNTH", | |
| 600 | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "ALPHA", | |
| "SYNTH", | |
| 600 | |
| ] | |
| ], | |
| "id": "positions", | |
| "label": "Positions preserve account scope and buy/sell direction", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [ | |
| [ | |
| "ORD-1001", | |
| "MISSING_CANCEL_ACK" | |
| ] | |
| ], | |
| "expected": [ | |
| [ | |
| "ORD-1001", | |
| "MISSING_CANCEL_ACK" | |
| ] | |
| ], | |
| "id": "exception_disposition", | |
| "label": "Required unresolved issues remain open; resolved cases are not unnecessarily escalated", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "exception_evidence", | |
| "label": "Escalations cite observed evidence from the correct order", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "exception_source_check", | |
| "label": "The required source was queried before declaring its evidence incomplete", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "response_provenance", | |
| "label": "Cancellation dispositions are supported by observed authoritative responses", | |
| "passed": true | |
| }, | |
| { | |
| "actual": [], | |
| "expected": [], | |
| "id": "safe_scope", | |
| "label": "No cross-scope or contradictory financial mutation was attempted", | |
| "passed": true | |
| }, | |
| { | |
| "actual": "ed4d85c6de424e4e48bd46a88a2e78970edd3c7c3d7c7bdc124c550ac36c2f95", | |
| "expected": "ed4d85c6de424e4e48bd46a88a2e78970edd3c7c3d7c7bdc124c550ac36c2f95", | |
| "id": "evidence_integrity", | |
| "label": "Authoritative source records remain immutable", | |
| "passed": true | |
| }, | |
| { | |
| "actual": true, | |
| "expected": true, | |
| "id": "evidence_provenance", | |
| "label": "Every booking refers to observed source evidence", | |
| "passed": true | |
| } | |
| ], | |
| "full_capture": "captures/escalated-full.png", | |
| "metrics": [ | |
| [ | |
| "600", | |
| "shares booked" | |
| ], | |
| [ | |
| "400", | |
| "shares unresolved" | |
| ], | |
| [ | |
| "1", | |
| "open evidence request" | |
| ], | |
| [ | |
| "14/14", | |
| "outcome checks passed" | |
| ] | |
| ], | |
| "model": "claude-sonnet-5-5", | |
| "run_id": "bb397ea01db84e39aa8cd630589a4e53", | |
| "scenario_id": "missing_cancel_ack", | |
| "summary": "Correctly escalated: available evidence is reconciled and the unresolved issue remains explicit.", | |
| "title": "Keep missing evidence unresolved." | |
| } | |
| ] | |
| </script> | |
| </body> | |
| </html> | |