Skip to content

v0.0.6

Choose a tag to compare

View all tags
@github-actions github-actions released this 03 May 18:28
· 212 commits to main since this release
v0.0.6
d29b76a

Install

Pick your platform and download:

Platform Asset
macOS Apple Silicon unbrowser-aarch64-apple-darwin.tar.gz
macOS Intel unbrowser-x86_64-apple-darwin.tar.gz
Linux x86_64 (glibc) unbrowser-x86_64-unknown-linux-gnu.tar.gz

One-liner (macOS / Linux):

curl -L https://github.com/protostatis/unbrowser/releases/download/v0.0.6/unbrowser-$(uname -m | sed 's/arm64/aarch64/')-$(uname -s | tr A-Z a-z | sed 's/darwin/apple-darwin/;s/linux/unknown-linux-gnu/').tar.gz \
  | tar xz -C /usr/local/bin
unbrowser --mcp

Verify checksums:

shasum -a 256 -c unbrowser-*.tar.gz.sha256

What's new

See auto-generated changelog below.

What's Changed

  • policy: MVP Tier-1 deterministic blocklist by @protostatis in #2
  • phase-a: observability — navigation_id, script_decision, policy_trace, report_outcome by @protostatis in #4
  • navigate: auto-extract on SPA-shell pages with embedded JSON by @protostatis in #3
  • train: T1 corpus collection harness for the prefit pipeline by @protostatis in #5
  • dom: dynamic <script src> insertion → fetch + eval + load event by @protostatis in #6
  • network_store: capture content-bearing fetch responses for extraction by @protostatis in #7
  • shims: content-positive observers — IntersectionObserver, ResizeObserver, MutationObserver by @protostatis in #8
  • settle: structured exit reason {idle, budget_exhausted, max_iters} by @protostatis in #9
  • docs: probabilistic execution policy design by @protostatis in #1
  • dom: HTML fragment parsing — innerHTML setter + insertAdjacentHTML by @protostatis in #10
  • modules: best-effort ES module loader (static + dynamic <script type=module>) by @protostatis in #11
  • bytecode-cache: skip QuickJS parse on repeat visits by @protostatis in #12
  • prefit: T2 + T3 + R1 + R2 — end-to-end prefit pipeline + v0 hand-curated bundle by @protostatis in #13
  • prefit pipeline, challenge auto-solve, settle improvements by @protostatis in #14
  • prefit: Bayesian aggregator + Thompson sampling at runtime by @protostatis in #17
  • outcomes: derive synthetic success/failure per navigate by @protostatis in #15
  • train: 100-site corpus + parallel collector by @protostatis in #16

New Contributors

Full Changelog: v0.0.5...v0.0.6

Contributors

protostatis
Assets 8
Loading