Releases: tripledoublev/v100
Releases · tripledoublev/v100
v0.2.5
Full Changelog: v0.2.4...v0.2.5
v0.2.4
Full Changelog: v0.2.3...v0.2.4
v0.2.3
Full Changelog: v0.2.2...v0.2.3
Full Changelog: v0.2.2...v0.2.3
Full Changelog: v0.2.2...v0.2.3
Full Changelog: v0.2.2...v0.2.3
v0.2.2
Full Changelog: v0.2.1...v0.2.2
v0.2.1 — Strategic Research Alignment
Finalized Phase 100 implementation and aligned Phase 200 roadmap with frontier agentic research goals.
Full Changelog: v0.2.0...v0.2.1
v0.2.0 — Phase 100: Recursive Self-Evolution
This release introduces the first milestone of the self-evolution engine, allowing agents to distill their own trajectories and author new tools at runtime.
Self-Evolution Engine
- Distill command —
v100 distill <run_id>converts JSONL traces into ShareGPT-formatted datasets for model fine-tuning and DPO. - Dynamic Tool Registry — support for
RegisterAndEnableat runtime, enabling agents to expand harness capabilities without re-compilation. - Automatic Build Feedback — modified
internal/core/loop.goto triggergo build ./...after every workspace mutation, injecting compiler errors as aSYSTEM ALERTto enforce a reality-check loop.
Dynamic Tools
sql_search— Execute SQL queries against local SQLite databases with path sanitization.graphviz— Render DOT graph definitions into images (PNG/SVG) for architectural visualization.
Improvements
- Dependency Tracking — Added
github.com/mattn/go-sqlite3for local structured data operations. - Documentation — New DF-11 quest in
dogfood/for verifying self-evolution trajectories.
Full Changelog: v0.0.3...v0.2.0
v0.1.0
Full Changelog: v0.0.3...v0.1.0
v0.0.3
Full Changelog: v0.0.2...v0.0.3
v0.0.2
Full Changelog: https://github.com/tripledoublev/v100/commits/v0.0.2