Skip to content

v0.2.0

Choose a tag to compare

View all tags
@haoranc haoranc released this 21 Feb 08:27
· 13 commits to main since this release
v0.2.0
8a1cc5a
This commit was created on GitHub.com and signed with GitHub’s verified signature.
GPG key ID: B5690EEEBB952194
Verified
Learn about vigilant mode.

What's New

Changed

  • Review overhead model is now additive (0m / 15m / 25m) instead of percentage-based. ReviewMode values: none, standard, complex. Legacy self and 2x-lgtm still accepted for backwards compatibility. (#46)

Added

  • Tier auto-correction heuristics: auto-upgrades to L when scope signals exceed thresholds (tests > 20, lines > 200, concerns >= 3) and auto-downgrades to XS for trivial tasks. --no-auto-tier flag to disable. (#47)
  • Co-dispatch warm context: when 2+ tasks target the same agent in one wave, auto-applies 0.5x warm context duration reduction to tasks beyond the first. (#48)
  • Modifier product floor of 0.10 to prevent sub-10m pathology when modifiers stack aggressively. (#50)
  • Batch wave estimation: amortizes review overhead across same-agent tasks per wave. (#49)
  • Non-coding task type estimation: --type flag for brainstorm, research, config, and documentation tasks with category-specific models. (#55)
  • Multi-agent session estimation: agent-estimate session subcommand for coordinated workflows. (#56)

Fixed

  • 11 post-LGTM nits: fractional minute rounding, deterministic wave tie-breaking, tightened keyword patterns, parallel efficiency calculation, and JSON report completeness. (#63)

Full Changelog: v0.1.0...v0.2.0

Assets 2
Loading