skepsun/dsh-engram

README 摘要

dsh-engram English · 中文 Minimalist long-term memory for DeepSeek Harness, distilled from the symbolic-index and pi-esr ideas — with one goal: save tokens . - Zero-LLM intake — auto-captures meaningful events from tool results by pure pattern matching (git milestones with a written -m commit message, edits to key files, repeated errors), plus an explicit engram store . Nothing on the hot path calls a model, and pure plumbing — git push / git stash / silent commits — is deliberately never recorded (see Auto-capture policy below). - Symbolic index + progressive disclosure — a compact [ENGRAM] block (default budget 700 chars ≈ 175 tokens; one line per memory) is injected at prompt assembly and frozen per session , keeping the request prefix byte-stable for KV-cache reuse. The agent drills down with engram recall / engram detail instead of dumping raw hits into context. - ESR-lite closure protocol — esr task / esr close / esr link give tasks a draft → active → stable lifecycle where stable requires real evidence ( artifact / evaluation / memory ref ), surfacing closure gaps instead of letting the agent declare victory without proof. - Memory GC (pi-esr constraints) — a scheduled, mechan…

分类