chenzheshushi-commits/dsh-evolve0

dsh-evolve

Self-evolving memory + skill lifecycle for DeepSeek Harness. Cross-session memory with zero-token deterministic recall (bigram-Jaccard fused with FTS5 BM25 via RRF), a tiered approval gate, and reinforcement that strengthens what you repeat. Procedural knowledge crystallizes into SKILL.md files that refine in place and are curated through an active-stale-archived lifecycle (reversible archive, pre-op backups, rollback, never deletes). Includes background per-turn review, anti-bloat convergence for both skills and memory, an auto-grown user profile, and a web settings page.

包名
dsh-evolve
版本
0.4.1
许可证
MIT
最近更新
2026年8月23日

安装

$npx -p @deepseek-ai/dsh dsh plugin --profile web add github:chenzheshushi-commits/dsh-evolve

Configuration

Everything is configurable through the plugin's settings page (web profile) or your DSH config. Notable switches:

KeyDefaultEffect
autoConfirmEnabledtruefalse = every model write waits for review
reviewEnabledtrueBackground per-turn review
reviewEveryTurns5Review throttle
reviewModel(main model)Route review to a different model
refineLLMfalseUse an LLM pass when crystallizing/refining skills
reinforceEvery3Observations per importance step
memoryMaxChars20000Memory character budget (0 disables)
convergeSuggesttrueSurface merge/fold suggestions
curatorStaleDays / curatorArchiveDays30 / 60Skill lifecycle thresholds
ftsEnabledtruefalse = pure bigram recall, no SQLite

The LLM is only ever used for optional auxiliary passes — skill refinement, background review, and skill merging. All of them are single-shot, skippable, and fall back to deterministic behavior on failure. Nothing runs in your main loop.