Skip to content

v0.7.0

Choose a tag to compare

@github-actions github-actions released this 24 Jul 16:12
· 3421 commits to main since this release
d7f4330

Open Science v0.7.0

A feature release: a live context-usage indicator, a streamlined first-run onboarding flow, native Office document previews (DOCX/XLSX/PPTX) isolated in their own processes, zoom & pan in the image viewer, a redesigned permission dialog with a code preview and per-grant scope, and the StepFun Step Plan subscription vendor — plus a round of notebook, ACP, and Windows fixes on top of v0.6.1.

Open Science is an open-source, model-agnostic AI workbench for scientific discovery — a self-hosted desktop app pairing a plan-and-execute agent with a persistent, sandboxed compute kernel and durable project/session storage.

This is a minor release: no breaking changes. It sharpens the desktop workspace (context usage, activity grouping, dialog chrome), expands file previews to Office documents and zoomable images, redesigns the tool-approval dialog around explicit grant scopes, reworks first-run onboarding into focused steps, and adds a new subscription vendor.

✨ Highlights

  • Live context-usage indicator. See how much of the model's context window a session is using, right in the workspace. (#409)
  • Native Office document previews. Preview DOCX, XLSX, and PPTX files inline, rendered in isolated out-of-process frames so a document can't destabilize the app. (#394)
  • Zoom & pan in the image viewer. High-resolution scientific figures open with Ctrl/Cmd+wheel / pinch zoom (1×–8×), drag-to-pan, and double-click to reset — with reduced-motion respected. (#374, #338)
  • Redesigned permission dialog. The approval prompt now shows a syntax-highlighted code preview and lets you pick the grant scope — this call only, or the whole conversation — with hardened notebook/shell tool identification. (#381, #403)
  • Streamlined first-run onboarding. Setup is split into focused environment, agent, provider, notebook, and location steps; environment repair moves into Settings. (#359)
  • StepFun Step Plan vendor. Add the Step Plan quota subscription as a built-in vendor over both the Anthropic and OpenAI-compatible protocols. (#411)

🚀 New Features

  • Context-usage indicator — a live readout of context-window consumption for the active session. (#409)
  • Office document previews — DOCX, XLSX, and PPTX rendered inline and full-screen, isolated in OOPIFs, with a download action for unsupported files. (#394)
  • Image viewer zoom & pan — Ctrl/Cmd+wheel / trackpad-pinch zoom, drag-to-pan, double-click reset, and an on-screen zoom control, honoring prefers-reduced-motion. (#374, #338)
  • Permission dialog with scope controls — a collapsible code preview plus per-grant scope (Once / This conversation), canonical option labels, and cross-framework notebook-cell matching. (#381)
  • Activity groups with declared purpose titles — the agent declares a title before each group of tool calls, so the transcript reads as labeled units of work. (#387)
  • Remembered close-action preference — the close/quit-during-a-run confirmation can remember your choice. (#388)
  • StepFun Step Plan — a new built-in subscription vendor (step-3.7-flash multimodal, plus step-3.5-flash / step-3.5-flash-2603 / step-router-v1) routed under /step_plan. (#411)

🔧 Improvements

  • Unified dialog chrome and transitions across every modal and popover, with the approval card restyled to match. (#372, #403)
  • Regenerated the Windows .ico to the white rounded-tile mark so the packaged exe, installer, and pinned icons match icon.png/icon.icns. (#395)
  • AI code-review pipeline reworked — dual Codex reviewers with independent backends, added review telemetry and parallel analysis, simplified check display names, and reviewers restricted to static, read-only inspection. (#401, #392, #407, #386)

🐛 Bug Fixes

  • Notebook managed cache (Windows) — create package caches with a protected DACL before writing trust markers so inherited drive/profile permissions can't reject every candidate, and fit the cache under long Windows profile paths (MAX_PATH), with per-candidate rejection diagnostics. (#406, #397)
  • Session isolation across framework switches — pin bridged Codex runtimes to immutable targets and scope connection state per session so a provider/credential change can't retarget or misclassify older sessions; close runtime-owned MCP hosts on teardown. (#376)
  • Cross-framework notebook cells — detect notebook MCP tools in both the hyphenated (Claude Code) and underscore (Codex/gpt bridge) server-name forms so gpt notebook cells render as code, and default tool rows to collapsed. (#385)
  • Workspace regression — restore workspace behavior lost in the permission-dialog update. (#393)
  • Artifact sources — accept a nested artifact source object serialized as a JSON string, while keeping strict union validation for genuinely malformed sources. (#390)

📦 Install

Requirements: macOS 12+ (Apple Silicon or Intel), Linux x64, or Windows 10/11 x64. On first run the app's onboarding wizard checks your environment and can install and configure a Claude-compatible model backend for you. Once installed, the app updates itself in place.

Download the build for your platform from the Assets below:

Platform File
macOS (Apple Silicon) aipoch-open-science-0.7.0-mac-arm64.dmg
macOS (Intel) aipoch-open-science-0.7.0-mac-x64.dmg
Linux aipoch-open-science-0.7.0-linux-x86_64.AppImage / aipoch-open-science_0.7.0_amd64.deb
Windows aipoch-open-science-0.7.0-win-x64-setup.exe

macOS — first launch. Release builds are Developer ID signed and notarized by Apple, so they open like any other app. (A .app you build yourself is not notarized — see README → Building From Source.)

Windows — first launch (unsigned build). No Authenticode certificate yet, so SmartScreen shows a bypassable "unrecognized app" prompt (More info → Run anyway).

Build from source instead:

npm install
npm run build:mac   # or: build:linux / build:win

🧭 What's in this release (maturity)

Aligned with the PRD and Roadmap. This is a minor release: it adds features within the existing phases — richer workspace signals (context usage, activity grouping), expanded file previews, a redesigned approval flow, and reworked onboarding — on top of v0.6.1. No Roadmap phase changes status.

  • ✅ Implemented: desktop shell (Electron + React + TS) with a tray/single-instance lifecycle, reachable as a desktop app, an optional localhost web UI (with project/session deep links and lifecycle sync), a headless backend mode, and an open-science CLI daemon plus a zero-dependency Node.js task SDK; ACP agent runtime on a pluggable framework backend (Claude Code default, with OpenCode and Codex as alternatives — Codex usable via an existing ChatGPT/Codex subscription login, and Claude via an isolated claude setup-token subscription) whose app-managed runtimes are installable, switchable, and uninstallable from Settings; tool-activity cards grouped under agent-declared purpose titles with a live context-usage indicator; an opt-in reviewer with a bounded fix loop (AI-review pipeline now dual-Codex, static-inspection-only); a persistent multi-kernel notebook runtime (warm Python/R/REPL kernels, app-managed conda environments with offline provisioning and bring-your-own interpreters, crash-recoverable run history, session .ipynb export); project + session persistence with a paginated, session-grouped Files tab and session pinning; artifacts MCP server; file renderers (incl. Office DOCX/XLSX/PPTX in isolated processes and a zoom/pan image viewer) + notebook/session-notebook viewers + a full-screen preview mode + offline molecule viewer; multi-provider config (built-in vendors Anthropic (Claude), DeepSeek, Zhipu AI (GLM), GLM Coding Plan, Kimi (Moonshot), MiniMax, StepFun, StepFun Step Plan, Xiaomi MIMO, SenseNova, Volcengine Ark, plus the OpenRouter aggregation gateway, custom gateways, and Claude/Codex subscription logins) with per-model image-input metadata and per-model reasoning-effort control; desktop task notifications; remote SSH compute hosts as notebook execution targets; resilient range-resume downloads; signed-and-notarized macOS packaging plus Windows/Linux, with in-place auto-update and a configurable data-storage location; file-based agent skills; 24 built-in connectors plus custom MCP servers; a streamlined guided onboarding wizard (focused environment/agent/provider/notebook/location steps); and per-conversation approval profiles with a scoped, code-previewing permission dialog.
  • 🚧 Partial: R remains managed-only and environment snapshotting/export is not built; the CLI/SDK surface is a first entry point and task-automation layer, not yet a fully scriptable/embeddable orchestration SDK; provider choice is still gated on the selected framework's endpoint compatibility rather than one unified gateway; remote compute covers SSH hosts for notebook runs but not Slurm/cloud-GPU submission or sub-agent fan-out; skills are local and file-based (no shared/public commons yet); the reviewer verifies a turn's own record but does not yet apply domain-specific citation/units/statistics rubrics.
  • 🗺️ Roadmap (not in this build): public skills commons and savable specialist roles (rest of Phase 3); a fully model-agnostic gateway with unified per-agent routing and a full CLI/SDK entry point (rest of Phase 1); artifact versioning + provenance chain, environment snapshotting/export, and specialist sub-agents (rest of Phase 2); Slurm/cloud-GPU remote compute, sub-agent fan-out, and the rest of the security stack — network sandbox, credential vault, scoped permission tiers (rest of Phase 4).

🐢 Known Limitations

  • R is managed-only. A bring-your-own R interpreter path is not built yet.
  • No environment snapshotting or export yet. Phase 2 work.
  • Remote compute is SSH-only. Slurm, cloud GPU, and sub-agent fan-out are not built yet.
  • Provider choice is per backend, not one unified gateway. Claude Code reaches Anthropic-compatible endpoints; OpenCode and Codex add OpenAI-compatible / Responses ones.
  • The task SDK is a first-generation surface. Task creation, polling, and artifact fetch work; a broader programmatic orchestration API is still open Phase 1 work.
  • Switching backends can't carry a live session over. Open conversations replay to the new backend, but in-flight tool state is not migrated.
  • Skills are local only. No shared/public commons, cross-machine forking, or user-facing version pinning yet.
  • The reviewer is opt-in and record-scoped.
  • Windows builds are unsigned. SmartScreen prompt on first launch. (macOS is notarized.)
  • No provenance chain / artifact versioning yet. Phase 2.
  • No local GPU compute backend.
  • No multi-user / real-time collaboration.

🙏 Acknowledgements

Thanks to @ewen-poch, @roxi3906, @LJLJ111, @imjszhang, @wen2zhou, @GiaoLee, and everyone in Discord, X, and Discussions.


What's Changed

  • fix(workspace): detect notebook cells across frameworks; collapse tool rows by @ewen-poch in #385
  • fix(acp): isolate sessions across framework switches by @ewen-poch in #376
  • ci(review): restrict AI reviewers to static inspection by @ewen-poch in #386
  • feat(window): remember close action preference by @ewen-poch in #388
  • feat(preview): zoom & pan in the image viewer (#338) by @roxi3906 in #374
  • fix(artifacts): accept stringified source objects by @ewen-poch in #390
  • feat(workspace): show run code and scope picker in permission dialog by @ewen-poch in #381
  • feat(activity-groups): declare titles before tool groups by @ewen-poch in #387
  • fix(workspace): restore behavior lost in permission update by @ewen-poch in #393
  • fix(branding): regenerate Windows icon.ico to white-tile variant by @ewen-poch in #395
  • ci(review): add AI review telemetry and parallel analysis by @ewen-poch in #392
  • feat: unify renderer dialog chrome and transitions by @LJLJ111 in #372
  • fix(notebook): fit managed cache under long Windows profiles by @ewen-poch in #397
  • feat(preview): isolate Office previews in OOPIFs by @roxi3906 in #394
  • feat(onboarding): streamline first-run onboarding flow by @roxi3906 in #359
  • fix(notebook): harden managed cache ACLs by @ewen-poch in #406
  • ci(review): replace Claude with dual Codex reviews by @ewen-poch in #401
  • ci(review): simplify check display names by @ewen-poch in #407
  • style(permission): align approval card with dialog chrome by @ewen-poch in #403
  • feat(acp): add context usage indicator by @ewen-poch in #409
  • feat(providers): add StepFun Step Plan vendor by @ewen-poch in #411
  • chore(release): bump version to 0.7.0 by @ewen-poch in #413

New Contributors

Full Changelog: v0.6.1...v0.7.0