xiaoshuai
a64ea2acb3
A durable memory layer for OpenCode. Structure:
ROOT.md total plan: goals, hard rules, directory registry,
and a scenario -> file routing table
<category>/
index.md one line per file, COMPILED from frontmatter
<slug>.md the memory itself
Two roots, same shape: <config>/memory/ (global) and
<project canonical>/.memory/ (project). Project wins on conflict; global
hard rules stay resident everywhere.
Why a plugin rather than a skill that tells the model to read a folder:
model-driven retrieval is unreliable — the model forgets, reads half the
files, or reads the wrong ones. Markdown is the human-facing format; the
plugin is the engine that parses, scores, and injects.
Measured, not assumed: FTS5's two built-in tokenizers both fail on this
corpus. On SQLite 3.53.4 with "部署脚本用 pnpm 而不是 npm...": unicode61
emits 7 terms for the whole sentence and matches 0 hits for "部署";
trigram needs >= 3 characters and also returns 0 for that two-character
word. So tokenization happens in the plugin (CJK bigrams, latin words)
and no SQL index is used — a thousand memories are fine to scan.
index.md and ROOT's two tables are compiled, never hand-edited: a
hand-maintained index drifts, a generated one cannot. Hit counters live
in .state.json so cross-device sync does not conflict on every recall.
Tools are named recall/remember/forget/inspect rather than
write/read/status: generic names were observed shadowing the harness's
own built-in tools.
1.1 KiB
1.1 KiB
id, kind, summary, highlight, created, updated, load_when
| id | kind | summary | highlight | created | updated | load_when | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| decisions/memory-store-format.md | fact | 选 Markdown 因为人类可纠正性和可同步性，检索由插件做而非模型。 | 纯文本可手改、可 git diff；SQLite 的相关性排序解决的是规模问题，而记忆只有千级 | 1791512156858 | 1791512156858 |
|
评估过 SQLite + FTS5，最终选 Markdown + 插件检索。
为什么放弃 SQLite：
- 模型自动抽取一定会抽错。没有人类可读可编辑的存储，就没有纠正入口。记忆错了比没有记忆更糟。
- FTS5 的中文分词两种内置方案都不可用（见 pitfalls/sqlite-cjk-tokenizer 那条实测）。
- 相关性排序是规模问题，而记忆是千级条目，解决的是不存在的问题。
Markdown 的代价及对策：
- 检索不可靠 → 由插件解析、评分、注入，不依赖模型自觉
- 写入抖动（hit_count 每次检索都改文件）→ 统计拆到 .state.json，git 忽略
已知边界：万条量级时全表扫描会变慢，那时再换索引（正文是纯文本，迁移成本低）。