Repository Issue Activity (beta)

#494 Fix static JSON evaluator parsing for noisy final numeric answers

#12 Dependency Dashboard

#297 remove tsfm specific csv from scenarios, reloate the files to couchdb

#530 Optional deterministic, unit-testable ground truth for rule-shaped scenarios (Tln)

#452 Helpter Notebook to Understand MCP Tools and Agents

#475 Static JSON evaluator fails noisy final MCQ answers

#398 Bring Model Store and Feature Store for Anomaly Detection

#470 Ensure Benchmark is completed with Stirrup Agent

#502 Industrial Automation Challenge Track 1 grader format clarification

#465 Experimental - result extraction

#527 Bring SubAgent into Echo System

#461 [Proposal] Physics-Based Anomaly Detection (AD) MCP Server for ICS Security Scenarios

#462 tsfm mcp redesign

#524 Portable interchange for AssetOpsBench ground-truth trajectories (EvalPort) — useful alongside the HF dataset?

#453 Preserve Stirrup per-scenario code workspaces for benchmark verification

#455 Stirrup scenario-suite runs still hit max context/token limit even after lowering max tokens

#457 opencode-agent fails with tokenrouter/x-ai/grok-4.5 due to schema validation error

#522 StirrupAgentRunner's default model id never matches a price-table key, so est_cost silently returns None

#523 Clean Open Scenarios for Version 2.0

#517 Code Change based on AA feedback

#515 Bare-name tool alias silently re-registers on a 3rd+ collision, only the 2nd logs a warning

#360 Investigate adaptive escalation policy for Plan-Execute orchestration

#511 An empty completion crashes `plan_execute` with a `TypeError` that blames the regex

#508 Two scenarios are answerable with no tools mounted — a contamination control, and what it caught

#509 None of the 2,667 FailureSensorIQ questions reference an entity in the loaded database