All pull requests

Pull requests

• fix(scenarios, simulator): red-team floor, rare quiet lines, unique names, coherence, caller STT and confirmations

  #140·KarthikAvinashFI opened Oct 1, 2026Member·

  ·
  

• fix(harness): require everything=true before dropping every scenario

  #128·KarthikAvinashFI opened Sep 30, 2026Member·

  ·
  

• fix(harness): apply an edited max_turns to the scenario a run plays

  #126·KarthikAvinashFI opened Sep 30, 2026Member·

  ·
  

• feat(harness): scenario suite quality and scale

  #101·KarthikAvinashFI opened Sep 21, 2026Member·

  ·
  

• feat(harness): gate the scenario faults that make a clean audit lie

  #98·KarthikAvinashFI opened Sep 21, 2026Member·

  ·
  

• feat(harness): let the main loop delegate scenario writing to its own sub-agents

  #95·KarthikAvinashFI opened Sep 17, 2026Member·

  ·
  

• Derive a coverage grid from the agent's contract and sample it for any suite size

  #75·KarthikAvinashFI opened Sep 2, 2026Member·

  ·
  

• Derive a coverage grid before writing scenarios, and sample it deliberately

  #73·KarthikAvinashFI opened Sep 1, 2026Member·

  ·
  

• Generalise the harness and close the defects that made runs look verified

  #70·KarthikAvinashFI opened Aug 30, 2026Member·

  ·
  

• Harden hosted environment resolution and reporting

  #68·hadarishav opened Aug 26, 2026Contributor·

  ·

• Hosted snapshot: drop Docker (A1), rabbitmq mgmt present-not-enabled (A3), guarantee /run/futureagi + /work layout (P5)

  #66·azain-commits opened Aug 25, 2026Member·

  ·
  

• feat(harness): hosted execution guest — provisioner, scheduler, outbound evidence, entrypoint

  #65·khushalsonawat opened Aug 25, 2026Member·

  ·
  

• Generate a suite with one writer per use case, and persona values the platform understands

  #63·KarthikAvinashFI opened Aug 22, 2026Member·

  ·
  3

  

• fix(evals): StepEfficiency negative score on redundant calls within a step

  #62·usegitdivyansh opened Aug 21, 2026·

  ·
  1

  4

• feat(simulate): sip_inbound_trunk_id transport field for carrier pinning

  #59·azain-commits opened Aug 17, 2026Member·

  ·
  1

  

• feat(harness): agent source registry and the understand-agent stage

  #58·KarthikAvinashFI opened Aug 16, 2026Member·

  ·
  

• feat(generation): grounded scenario generation with deterministic checks and live voice wiring

  #57·KarthikAvinashFI opened Aug 15, 2026Member·

  ·
  

• fix(protect): align metric names across Python and TypeScript SDKs

  #56·KarthikAvinashFI opened Aug 6, 2026Member·

  ·
  

• fix(simulate): reclaim orphaned SIP dispatch rules instead of deadlocking

  #54·hadarishav opened Aug 1, 2026Contributor·

  ·

• feat(feedback): enrich ChromaFeedbackStore with torchembed metadata embeddings

  #52·py-ai-dev opened Jul 8, 2026·1 of 3 tasks·

  ·