Skip to content

Releases: adsrz/learning-os

Learning OS v0.1.1

03 Apr 06:53
@adsrz adsrz
v0.1.1
60a71f7

Choose a tag to compare

View all tags
Learning OS v0.1.1 Latest
Latest

Second public release of Learning OS, focused on making the post-v0.1.0 public direction explicit and packaged.

Highlights

  • community-entry surfaces: CONTRIBUTING.md, ROADMAP.md, and issue templates
  • explicit prompt-only versus harness comparison in the top README surfaces
  • visible command-proof matrix for the verified pwsh path and Windows convenience wrapper
  • a first clearly non-finance public-safe example: policy-brief-sample.md plus examples/thesis-reading-packet

Why this release exists

  • v0.1.0 proved the harness shape
  • this release packages the next public-facing step after v0.1.0
  • the repo now has clearer participation surfaces, clearer proof surfaces, and a stronger example mix

Included surfaces

  • community entry: CONTRIBUTING.md, ROADMAP.md, .github/ISSUE_TEMPLATE/
  • comparison and proof: README.md, README.zh-CN.md, docs/public-setup.md, docs/demo-flow.md
  • non-finance thesis-reading example: samples/open/policy-brief-sample.md, examples/thesis-reading-packet/, docs/examples/thesis-reading.md

Notes

  • no change to the BYOS boundary
  • no maintainer-only state inside the tracked public repo
  • this release keeps the public surface bounded rather than expanding into many new onboarding lanes at once
Assets 2
Loading

Learning OS v0.1.0

03 Apr 03:57
@adsrz adsrz
v0.1.0
This tag was signed with the committer’s verified signature.
adsrz
GPG key ID: 1FCA37BA5C5318C2
Verified
Learn about vigilant mode.
7c34f99
This commit was signed with the committer’s verified signature.
adsrz
GPG key ID: 1FCA37BA5C5318C2
Verified
Learn about vigilant mode.

Choose a tag to compare

View all tags
Learning OS v0.1.0

First public release of Learning OS as an AI-native, local-first learning harness.

Highlights

  • result-first public README with a visible 60-second proof path
  • worked public-safe packets for research, single-book, and multi-book flows
  • machine-readable agent entrypoints and routing contracts
  • repo-only validation for clean clones on Windows and pwsh
  • pwsh-first documented command path, with Windows .cmd wrappers kept as convenience entrypoints

Experience reflected in this release

  • proof before architecture: show a concrete packet outcome before asking new visitors to read the deeper harness split
  • portable commands only when verified: promote the pwsh path only after the actual script entrypoint has been tested locally
  • one release-facing validation signal: keep one public repo-only workflow instead of stacking near-duplicate CI surfaces
  • public result, private maintainer state: keep release working memory and maintainer routing outside the tracked public repo

Included surfaces

  • shared layer: AI_CONTEXT.md, ai-context.json, AGENTS.md, system.md
  • routing and write-back contracts: task-router.json, writeback-map.json
  • public-safe skills under agent/skills/
  • worked examples and open samples
  • validation and BYOS import helpers under tools/

Notes

  • internal values such as 1.3 in ai-context.json, task-router.json, writeback-map.json, and agent/skills/index.json are contract/schema versions, not repo semver
  • this release intentionally does not include third-party books, papers, PDFs, slides, or proprietary study materials
Loading