Issues

• [HOSTED-EXTRACTION-WORKER] Railway service for the extraction queue (blocked by #62)

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#362 · Filter by author wilsonhj opened on Oct 29, 2026

• [PUBLIC-AUTH/BYOK] Public signup, private workspaces and user-funded inference

  agent-taskcontract-change
  wilsonhj/Agentic-Financial-Evidence-Lab#324 · Filter by author wilsonhj opened on Sep 16, 2026
  1

• [AUTH] Register production token verification separately from authorized mock-auth smoke

  agent-taskcontract-change
  wilsonhj/Agentic-Financial-Evidence-Lab#292 · Filter by author wilsonhj opened on Sep 13, 2026
  1

• [CI/OBS] Second CODEOWNER and hosted Sentry verification remain (coverage floors, required checks, Sentry init and browser cache landed in #269)

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#203 · Filter by author wilsonhj opened on Sep 1, 2026
  2

• [EVALS] Retrieval release gates are a library, not a gate: nothing grades the benchmark seed against a built index in CI

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#201 · Filter by author wilsonhj opened on Sep 1, 2026
  3

• [ADR] Provider direction: ADR-0012 still Proposed; OpenAI adapters (#328/#329) exist but are not runtime-wired; no Anthropic adapter

  agent-taskarchitecturecontract-change
  wilsonhj/Agentic-Financial-Evidence-Lab#195 · Filter by author wilsonhj opened on Sep 1, 2026
  2

• [API] Live provider usage and cost accounting acceptance remains (ceilings, rate limits and bounded reads landed in #232/#272)

  agent-taskbugsecurity
  wilsonhj/Agentic-Financial-Evidence-Lab#191 · Filter by author wilsonhj opened on Sep 1, 2026
  1

• [ARCH-REVIEW 2026-09-01] System design and implementation review: findings, fixes, and follow-ups

  agent-taskarchitecture
  wilsonhj/Agentic-Financial-Evidence-Lab#188 · Filter by author wilsonhj opened on Sep 1, 2026
  5

• [RELEASE-LIVE-CUTOVER] Mock-to-live cutover: provision credentials, select the embeddings provider by benchmark, and run the live evaluation gates

  agent-taskcontract-change
  wilsonhj/Agentic-Financial-Evidence-Lab#177 · Filter by author wilsonhj opened on Aug 31, 2026
  3

• [M2-CLAIMS] Live 65-question exit gate (close M2-024)

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#132 · Filter by author wilsonhj opened on Jul 21, 2026
  1

• [READER-PROD-SMOKE] Hosted worker-to-browser production smoke (child F of #87)

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#108 · Filter by author wilsonhj opened on Jul 17, 2026
  4

• [READER-CROSS-STACK] Worker-to-browser reader verification (child E of #87)

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#96 · Filter by author wilsonhj opened on Jul 15, 2026
  6

  

• [M1-READER-INTEGRATION] Version-pinned production evidence reader

  agent-taskcontract-change
  wilsonhj/Agentic-Financial-Evidence-Lab#87 · Filter by author wilsonhj opened on Jul 14, 2026
  1

  

• [EXT-2b] Supplemental SEC stress cohort: restore the ≥8 distinct stress-feature bar

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#81 · Filter by author wilsonhj opened on Jul 13, 2026
  2

• [M5-AUDIT-RELEASE] Audit traversal, exports, release suites, frozen 300-question benchmark, gate verification, and the MVP release artifact

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#68 · Filter by author wilsonhj opened on Jul 12, 2026
  1

• [M5-BACKTEST] Rolling-origin backtests, calibration intervals, and the seasonal-naive gate for advanced models

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#67 · Filter by author wilsonhj opened on Jul 12, 2026

• [M5-FORECASTING] Forecast fit/predict/backtest interface, quarterly baselines, analyst driver forecasts, and Forecast Lab views

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#66 · Filter by author wilsonhj opened on Jul 12, 2026

• [M4-MODEL-UI] Interactive driver graph, analytics charts (bridges, waterfalls, heatmaps, tornado), and formula/lineage/diff/restore views

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#65 · Filter by author wilsonhj opened on Jul 12, 2026

• [M4-FACT-SCENARIOS] Source-backed model nodes from approved extractions and sparse bull/base/bear scenario layers

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#64 · Filter by author wilsonhj opened on Jul 12, 2026

• [M3-CONFIDENCE-GATE] Confidence calibration and review thresholds, monetary auto-approval prevention, and extraction release gates

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#62 · Filter by author wilsonhj opened on Jul 12, 2026
  1

• [M3-REVIEW] Accept/edit/reject/merge/rerun review workflows and versioned approved records with correction history

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#61 · Filter by author wilsonhj opened on Jul 12, 2026
  3

• [M1-CORPUS-QA] Tracker: T0111 suites merged in #86; live 20-issuer corpus artifact (T0112) is owned by #177

  agent-task
  wilsonhj/Agentic-Financial-Evidence-Lab#56 · Filter by author wilsonhj opened on Jul 12, 2026
  2

• [M5] Forecast Lab and release

  wilsonhj/Agentic-Financial-Evidence-Lab#8 · Filter by author wilsonhj opened on Jul 12, 2026

• [M4] Revenue and gross-profit model

  wilsonhj/Agentic-Financial-Evidence-Lab#7 · Filter by author wilsonhj opened on Jul 12, 2026

• [M3] Agentic extraction

  wilsonhj/Agentic-Financial-Evidence-Lab#6 · Filter by author wilsonhj opened on Jul 12, 2026