Skip to content

feat: add evidence-aware AutoResearch workbench - #2

Closed
redmaplewww wants to merge 4 commits into
mainfrom
codex/ar-006-research-workbench
Closed

redmaplewww wants to merge 4 commits into
mainfrom
codex/ar-006-research-workbench

Conversation

@redmaplewww

Copy link
Copy Markdown
Owner

Summary

  • add a responsive Next.js research workbench for idea intake, paper collection, reading, writing, and review
  • connect supported FastAPI boundaries while keeping E0 previews, browser-local drafts, blocked states, and backend Gate decisions distinct
  • add responsive/browser evidence, frontend tests, documentation, and synchronized Project-to-Act/VibeCollab ledgers

Verification

  • npm.cmd run lint
  • npm.cmd run typecheck
  • npm.cmd run test (3 passed)
  • npm.cmd run build
  • python -m ruff check src tests
  • python -m pytest -W error -q (26 passed)
  • autoresearch doctor (5 agents)
  • Project-to-Act validate
  • node .ai-team/check.mjs --base main
  • production browser at 1440x1000 and 390x844: 0 px horizontal overflow

Boundaries

  • local single-user Web slice only; authentication, collaboration, streaming events, production hosting, licensed corpus quality, and real MD results remain unverified
  • real local run returned blocked/waiting_evidence as expected; it is not presented as a successful research result

@redmaplewww

Copy link
Copy Markdown
Owner Author

Superseded by #3. PR #2 remains as frontend prototype evidence; AR-008 replaces its static/demo interactions with project-level routes, persisted /api/v1 contracts, explicit offline degradation, and verified customer journeys.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants