Issues

• Nous v2 follow-ups from the pre-merge branch review (non-blocking)

  AI-native-Systems-Research/agentic-strategy-evolution#332 · Filter by author mtoslalibu opened on Oct 23, 2026
  1

• s390x (IBM Z) deployment: update to Nous v2 CLI, campaign.yaml and artifact layout

  AI-native-Systems-Research/agentic-strategy-evolution#331 · Filter by author mtoslalibu opened on Oct 23, 2026

• Nous v2: one agent session, enforced validation, a light method, unattended resume

  AI-native-Systems-Research/agentic-strategy-evolution#330 · Filter by author mtoslalibu opened on Oct 21, 2026
  1

• Continue gate recommends 'continue' even when its own summary says the question is fully answered

  enhancementtokens
  AI-native-Systems-Research/agentic-strategy-evolution#329 · Filter by author mtoslalibu opened on Oct 20, 2026

• findings.json with any extra top-level key fails the iteration permanently and drops its principles (no repair attempt)

  bugquality
  AI-native-Systems-Research/agentic-strategy-evolution#328 · Filter by author mtoslalibu opened on Oct 20, 2026

• SDK dispatcher: 1 MB default max_buffer_size kills an iteration when one tool result is >1 MB (e.g. reading a PDF)

  bug
  AI-native-Systems-Research/agentic-strategy-evolution#327 · Filter by author mtoslalibu opened on Oct 20, 2026

• optimization: progress.json eta_seconds is unreliable when row cost varies across levels

  AI-native-Systems-Research/agentic-strategy-evolution#325 · Filter by author sriumcp opened on Oct 18, 2026
  1

• optimization: seat the known_valid_baseline in multi-objective (Pareto) confirm rounds

  AI-native-Systems-Research/agentic-strategy-evolution#324 · Filter by author sriumcp opened on Oct 18, 2026
  1

• optimization: max_wall_clock_hours resets on nous resume (per invocation, not per campaign)

  AI-native-Systems-Research/agentic-strategy-evolution#323 · Filter by author sriumcp opened on Oct 18, 2026
  1

• optimization: survivor check fails the whole epoch on processes that were still exiting after a timeout kill

  AI-native-Systems-Research/agentic-strategy-evolution#322 · Filter by author sriumcp opened on Oct 18, 2026
  1

• optimization: deferred items from the #302 expert audit (tracking)

  AI-native-Systems-Research/agentic-strategy-evolution#321 · Filter by author sriumcp opened on Oct 17, 2026
  1

• optimization: decision-theoretic run allocation (WP-D; recombination pool, nous plan-next, optional Optuna planner)

  AI-native-Systems-Research/agentic-strategy-evolution#320 · Filter by author sriumcp opened on Oct 17, 2026
  1

• optimization: run_test_command has a hard-coded 900s timeout with no campaign override, and a killed test command is reported as '0 test(s)' with no mention of the budget

  bugfriction-reportkind: optimizationux
  AI-native-Systems-Research/agentic-strategy-evolution#319 · Filter by author sriumcp opened on Oct 10, 2026
  2

• optimization: an empty observable is diagnosed as a type mismatch, and a threshold-bound factor reads as 'not demonstrably live' because liveness probes only the base corner

  enhancementfriction-reportkind: optimizationux
  AI-native-Systems-Research/agentic-strategy-evolution#318 · Filter by author sriumcp opened on Oct 10, 2026
  2

• optimization: a native_test declared with a path prefix never matches an absolute-path test_command, and the guide calls test_command a shell command when it is shlex.split

  bugfriction-reportkind: optimizationux
  AI-native-Systems-Research/agentic-strategy-evolution#317 · Filter by author sriumcp opened on Oct 10, 2026
  2

• SDK ProcessError (exit 143) from a gateway stream-drop is misclassified as permanent, aborting the whole campaign

  AI-native-Systems-Research/agentic-strategy-evolution#315 · Filter by author mtoslalibu opened on Oct 6, 2026
  2

• Iteration #3 wrongly routed to OpenAI's endpoint

  AI-native-Systems-Research/agentic-strategy-evolution#304 · Filter by author oprince opened on Aug 24, 2026
  2

• Campaign-authored recommended_turn_silence_threshold_seconds deadlocks the runner (bundle scalar overrides DESIGN's 600s default; retries leak event loops)

  AI-native-Systems-Research/agentic-strategy-evolution#303 · Filter by author sriumcp opened on Aug 16, 2026
  2

• Wiki launders flawed campaign self-reports: no cross-campaign reconciliation, confidence/provenance stripped

  AI-native-Systems-Research/agentic-strategy-evolution#301 · Filter by author namasl opened on Jul 14, 2026

• Campaign conclusions are never verified against raw measurements before becoming durable principles

  AI-native-Systems-Research/agentic-strategy-evolution#300 · Filter by author namasl opened on Jul 14, 2026

• Design gate should auto-populate verified_parameters from locked_parameters instead of hard-failing on omissions

  AI-native-Systems-Research/agentic-strategy-evolution#298 · Filter by author namasl opened on Jul 13, 2026

• [bench] Sub-issue C: 11-metric selectable judge rubric

  AI-native-Systems-Research/agentic-strategy-evolution#295 · Filter by author Naima2002 opened on Jun 18, 2026

• index-wiki should use target_repo (GitHub slug) as registry project key

  AI-native-Systems-Research/agentic-strategy-evolution#294 · Filter by author susiejojo opened on Jun 16, 2026

• Cheaper cost defaults: execute_analyze effort=medium, max_iterations=3 (route both through defaults.yaml)

  AI-native-Systems-Research/agentic-strategy-evolution#288 · Filter by author sriumcp opened on Jun 12, 2026

• Benchmarking frameworks

  AI-native-Systems-Research/agentic-strategy-evolution#283 · Filter by author Naima2002 opened on Jun 12, 2026
  2