Give the system a directory: ask who knows, then ask them

A memory system that only answers from its own store is a filing cabinet,
not a partner. The transactive rule the project is named for — not knowing
is fine when you know who does — needed a holder registry, a directory
query, and a referral that never leaks the text it may not read.

- schema v5 adds `holders` (person/agent/service, declared topics, contact).
  A holder with memories exists whether or not anybody registered them; the
  registry adds reachability, it does not grant existence.
- `who_knows` ranks holders by their own memories and is deliberately NOT
  filtered by visibility: "she knows" is the answer. The answer is filtered.
- `ask` searches the asker's visible scope first (hop 0), then walks the
  directory with a visited set and a hop budget. Three terminal states:
  answered, referred, no_holder_knows. Nobody knowing is a result, not an error.
- Confidence is ordinal, not calibrated: relative strength, coverage, and
  declared topics. Every hit carries the raw numbers and the basis lines that
  produced the score, so a ranking can be argued with.
- Faces: HTTP (/api/holders, /api/who_knows, /api/ask), CLI (holders,
  holder-set, holder-rm, who-knows, ask), MCP (who_knows, ask, list_holders).

Four bugs found by running the thing, not by reading it:
- FTS5 refuses bm25() inside MAX(); the aggregate uses FTS5's own `rank` column.
- A CJK question is longer than the sentence that answers it, so the whole-run
  phrase matched nothing. The query now ORs adjacent bigram phrases; adjacency
  is kept, equal length is not. Latin path untouched, baseline unaffected.
- The hop budget counted the asker as a hop, so max_hops=1 asked nobody.
- A second, stricter readability test in _answered() disagreed with the SQL
  visibility clause and hid a memory the asker was allowed to read. The SQL
  clause is the only authority now.

A method named `list` shadowed the builtin inside its own class and made
mypy reject the return annotation; renamed rather than cast around it.

Proof: 155 pytest, 12 Go tests, ruff and mypy clean on touched files, pyright
ledger unchanged at 62, and scripts/mcp_write_smoke.sh drives two principals
on one SQLite file — the husband asks through MCP, gets `referred` with his
wife's contact channel, and her private sentence appears nowhere in the reply.
This commit is contained in:
ZacharyZhang-NY committed 2026-10-07 17:20:45 -04:00
1 parent 9fe13fffb3
commit dfc6cd4c28
19 files changed
+1403 -19

No files matched your search

+28 -11
View File
@@ -76,17 +76,34 @@
- [x] p0-3.6 Go 的 `search_atoms` / `compile_context` 改走 `POST /api/search`,删掉 Go 侧第二套分词器(`store.SearchAtoms` / `ftsQuery` / `tokenPattern`)——两面不可能再漂开,因为只剩一个排序实现
- [ ] p0-3.7 100k atom 下 p50 <100ms:`python -m bench.scale_check` → `bench/reports/SCALE.md`。基线是 p50 20.16s / 165,973 atoms(`bench/reports/BASELINE.md`)。**未测不写数**:机器负载 3.7–17.9 期间测出来的数不算数
- 验收:`pytest backend/tests/test_text_index.py backend/tests/test_search_api.py` 绿(18 条,真 FTS5 表 + TestClient):「钥匙」命中「我们的钥匙放在玄关抽屉里」、`search atoms` 命中 `func SearchAtoms(ctx context.Context)`、`forg` 命中 forgotten、全角 `Search Atoms` 命中、游标翻页无重复、supersede 后旧记忆不再应答;MCP 一致性由 `ely-eye-mcp/internal/mcpserver/search_tools_test.go`(3 条)+ `bash scripts/mcp_write_smoke.sh` PASS 证明;偏差:checklist 写的 `tests/test_retrieval_parity.py` 不存在,实际在 `backend/tests/` 与 Go 测试里
- [ ] **p0-4** 卡带变成可挂载作用域
- [ ] p0-4.1 `mount` / `unmount` / `switch` / `append` 四动词
- [ ] p0-4.2 检索按已挂载卡带过滤(作用域,不再是全局)
- [ ] p0-4.3 `kv_fingerprint`:model / tokenizer / chat template / adapter hash / rope / 量化 / 引擎版本
- [ ] p0-4.4 fingerprint 不匹配则拒绝挂载,错误信息点名差异字段
- 验收:`pytest tests/test_cartridge_scope.py tests/test_kv_fingerprint.py` 绿
- [ ] **p0-5** holder registry + who_knows
- [ ] p0-5.1 `holders` 表 + registry CRUD
- [ ] p0-5.2 `who_knows` 工具:返回 holder 与置信度,非空结果
- [ ] p0-5.3 转介:visited-set + 深度上限 + 终止态 `NO_HOLDER_KNOWS`
- [ ] p0-5.4 默认 answer-only 转介,不外泄原文
- [x] **p0-4** 卡带变成可挂载作用域
- [x] p0-4.1 `mount` / `unmount` / `switch` / `append` 四动词(`cartridge.py:CartridgeService`;schema v2→v3 新增 `cartridge_atoms` 成员表与 `cartridge_mounts` 挂载表,重开时自动重建——挂载是机器状态,不是数据事实)
- [x] p0-4.2 检索按已挂载卡带过滤(`retrieval.py:search` 默认作作域 = 当前挂载集,显式 `cartridge_ids` 覆盖挂载;什么都没挂时仍是全库,空库行为与以前一致)——偏差:`forget` 同时从所有卡带成员表里掉出,否则“忘了”还能从卡带里搜到
- [x] p0-4.3 `kv_fingerprint`(`kv_fingerprint.py`):model / tokenizer / chat template hash、adapter hash、RoPE 块(factor / original_max_position_embeddings / mrope_interleaved / mrope_section / partial_rotary_factor)、context 长度、引擎版本、指纹格式版本;读不到的项写进 `unverifiable_fields`,不留空
- [x] p0-4.4 fingerprint 不匹配则拒绝挂载,错误信息逐字段点名差异(`home_x was built for a different runtime; refusing to mount: rope: factor: cartridge 4.0 vs local 2.0; …`)
- 验收:`pytest backend/tests/test_cartridge_scope.py backend/tests/test_kv_fingerprint.py` 绿(26 条:16 作用域 + 10 指纹);端到端 `bash scripts/mcp_write_smoke.sh` → `PASS: remember, supersede, search, forget and cartridge mounting work through MCP`(挂载后 work 的记忆搜不到、上传后新写的记忆要 append 才能检索、unmount 后全库恢复、改 RoPE factor 后 `mount_cartridge` 报错含 `rope`);偏差:checklist 写的 `tests/` 不存在,实际在 `backend/tests/`
- 偏差记录:四动词同时有 HTTP(`/api/mounts`、`/api/cartridges/{id}/{mount,switch,unmount,append}`)、CLI(`ely-eye mount/mounts/unmount/append`)和 5 个 MCP 工具;`append` 重写 `atoms.parquet` + `sparse_index.bm25.zst` 并重算 DNA,文件先写、后写库,失败不会留半个卡带;**量化没进指纹**——量化是引擎启动参数(`profiles.py` 的 `--cache-type-k q4_0` 等),不在模型快照里,指纹只比快照真说得出的东西,否则就是拿不存在的字段做门禁
- 跑出来的两个真缺陷:`atoms.parquet` 根本导不出记忆(pyarrow 拒绝无子字段的 struct,`metadata={}` 炸掉整表 → 六个 JSON 列改存 JSON 文本,读取端 `load_cartridge_atoms` 解码);`materialize()` 先写成员表后写 cartridges 行,外键拒绝所有卡带
- `api.py` 超 500 行,卡带/挂载路由拆到 `api_cartridges.py`;那个模块不能用 `from __future__ import annotations`——FastAPI 会拿模块 globals 去解 `Depends(settings_dep)`,而它只是 `register()` 的参数,结果每条路由都要 `settings` 查询参数
- [x] **p0-5** holder registry + who_knows
- [x] p0-5.1 `holders` 表 + registry CRUD
- 验收:`pytest backend/tests/test_transactive.py -k "registry or kind or blank or holding" -q` 通过;schema v5 在旧库上补表。
- 事实:`migrations.py` SCHEMA_VERSION 4→5,`holders` 进 `_BASE_TABLES`(holder_id 主键、kind CHECK、knows_about_json、contact_json、created/updated)。CRUD 在 `holder_store.py`(`db.py` 已 498 行,逼近 500 上限,故另立模块)。删除保护:`holder 'x' still holds N memories; forget them before deleting the holder` → HTTP 409。
- 偏差:方法名 `list` 在类内遮蔽内置 `list`,mypy 判 `list[Holder]` 注解非法(`Function ... is not valid as a type`)→ 改名 `list_holders()`,不是加 cast 糊过去。
- [x] p0-5.2 `who_knows` 工具:返回 holder 与置信度,非空结果
- 验收:`pytest backend/tests/test_transactive.py -k "directory or declared" -q`;`bash scripts/mcp_write_smoke.sh` 里 who_knows("护照") 首位是 wife 且带 contact.channel=signal。
- 事实:目录**不按可见性过滤**("她知道"就是答案),回答才按可见性过滤。置信度是序数的:`0.60*相对得分 + 0.25*min(1, matched/3) + 0.15*声明过这个话题`,相对 = top_score / 候选集最高 top_score(避免 BM25 魔法常数:同一 query 在 3 文档库上量级 1e-7)。每条命中附 matched / top_score / latest_time 与 basis 说明。
- 根因修复:FTS5 的 `bm25()` 不能聚合(`MAX(-bm25())` + GROUP BY 报 `unable to use function bm25 in the requested context`),子查询/窗口形式也被拒 → 用 FTS5 自带的 `rank` 列(`-MAX(rank)`,实测与 `MAX(-bm25())` 相等,9.1786e-07)。
- 根因修复:`fts_query("钥匙在哪里")` 生成与问句等长的短语 → 0 命中(实测 钥匙 ✓ / 钥匙在哪里 ✗)。`text_index.py` 新增 `_cjk_phrase`:整段短语 OR 每个相邻二元组短语,相邻性保留(p0-3 的反精确测试仍成立),等长不再是条件。拉丁路径未动 ⇒ 冻结的 BASELINE 数字不受影响。
- [x] p0-5.3 转介:visited-set + 深度上限 + 终止态 `NO_HOLDER_KNOWS`
- 验收:`pytest backend/tests/test_transactive.py -k "walks or hop or nobody" -q`。
- 事实:`ask()` 先查 asker 可见范围(hop 0),再按 who_knows 命中依次跳转,`asked` 列表即 visited-set,不重复问同一 holder;`max_hops` 限 1..10,越界 `ValueError("ask needs max_hops between 1 and 10, got N")`;终止态 `answered` / `referred` / `no_holder_knows`(无人知道是结果,不是错误:`basis=["no holder has a memory that answers this"]`)。
- 根因修复:`if len(asked) >= max_hops: break` 把 asker 自己算作一跳 ⇒ `max_hops=1` 谁也没问就返回 no_holder_knows → 改 `len(asked) - 1 >= max_hops`,docstring 写明「max_hops 数的是 asker 之外被问的人」。
- [x] p0-5.4 默认 answer-only 转介,不外泄原文
- 验收:`bash scripts/mcp_write_smoke.sh` → `PASS: remember, supersede, search, forget, cartridge mounting and the transactive referral work through MCP`(两个进程、一个 SQLite:husband 通过 MCP `ask` 得到 `status=referred`、`answered_by=wife`、`answer` 为空且整份 JSON 里搜不到妻子的原文)。
- 事实:`include_text` 只对可读所有者(asker 本人或无主的 library 条目)显示原文;`referred` 只带 contact 渠道。
- 根因修复:`_answered()` 里 `readable = owner in (as_holder, LIBRARY)` 是 SQL 可见性子句之外的第二份、更严的规则,与 `visible_to=[husband]` 冲突(共享记忆被错误隐藏)→ 删掉这份重复判断,SQL 的 `as_holder` 可见性子句是唯一权威。
- 偏差:Go 侧 `limit`/`max_hops` 零值曾当真实数字发出,被 API 以 `greater_than` 拒 → 未设置的可选字段一律省略,由服务端定默认值。
- 验收:`pytest tests/test_transactive.py` 绿;钥匙测试场景通过
- [ ] **p0-6** 证明套件去自证
- [ ] p0-6.1 删 arena 答案入墨(`evals.py:1408-1409` 连接线、`evals.py:1417-1426` 白圈)
+4
View File
@@ -51,3 +51,7 @@ where = ["src"]
line-length = 100
target-version = "py312"
[tool.ruff.lint]
# Typer requires an option default to be a typer.Option call, which is what B008 forbids.
ignore = ["B008"]
+2
View File
@@ -11,6 +11,7 @@ from pydantic import BaseModel
from . import __version__
from .api_cartridges import register as register_cartridge_routes
from .api_holders import register as register_holder_routes
from .benchmarks import (
benchmark_prediction_report,
benchmark_registry_report,
@@ -246,6 +247,7 @@ def create_app() -> FastAPI:
raise HTTPException(status_code=422, detail=str(exc)) from exc
register_cartridge_routes(app, settings_dep)
register_holder_routes(app, settings_dep)
@app.get("/api/adapters", response_model=list[AdapterStatus])
def adapters(settings: Annotated[Settings, Depends(settings_dep)]) -> list[AdapterStatus]:
+118
View File
@@ -0,0 +1,118 @@
"""Holder registry, who-knows, and referral routes: the directory side of the store.
No `from __future__ import annotations` here: FastAPI reads these route signatures, and string
annotations would be resolved against module globals, where the injected settings_dep exists
only as a parameter of register().
"""
from collections.abc import Callable
from typing import Annotated
from fastapi import Depends, FastAPI, HTTPException
from .config import Settings
from .db import Database
from .holder_store import HolderStore
from .schemas import utc_now
from .transactive import TransactiveService
from .transactive_schemas import (
AskRequest,
Holder,
HolderUpsertRequest,
ReferralAnswer,
WhoKnowsReport,
WhoKnowsRequest,
)
def register(app: FastAPI, settings_dep: Callable[[], Settings]) -> None:
"""Register the transactive routes. The server owns the timestamps, the caller does not."""
def store(settings: Settings) -> HolderStore:
return HolderStore(Database(settings))
@app.get("/api/holders", response_model=list[Holder])
def holders(settings: Annotated[Settings, Depends(settings_dep)]) -> list[Holder]:
return store(settings).list_holders()
@app.post("/api/holders/{holder_id}", response_model=Holder)
def put_holder(
holder_id: str,
request: HolderUpsertRequest,
settings: Annotated[Settings, Depends(settings_dep)],
) -> Holder:
now = utc_now()
entry = Holder(
holder_id=holder_id,
created_at=now,
updated_at=now,
**request.model_dump(),
)
try:
return store(settings).upsert(entry)
except ValueError as exc:
raise HTTPException(status_code=422, detail=str(exc)) from exc
@app.get("/api/holders/{holder_id}", response_model=Holder)
def get_holder(
holder_id: str,
settings: Annotated[Settings, Depends(settings_dep)],
) -> Holder:
entry = store(settings).get(holder_id)
if entry is None:
raise HTTPException(status_code=404, detail=f"no holder {holder_id!r} is registered")
return entry
@app.delete("/api/holders/{holder_id}", response_model=Holder)
def delete_holder(
holder_id: str,
settings: Annotated[Settings, Depends(settings_dep)],
) -> Holder:
registry = store(settings)
entry = registry.get(holder_id)
if entry is None:
raise HTTPException(status_code=404, detail=f"no holder {holder_id!r} is registered")
try:
registry.delete(holder_id)
except ValueError as exc:
raise HTTPException(status_code=409, detail=str(exc)) from exc
return entry
@app.get("/api/holders/{holder_id}/knows", response_model=WhoKnowsReport)
def holder_knows(
holder_id: str,
query: str,
settings: Annotated[Settings, Depends(settings_dep)],
) -> WhoKnowsReport:
"""What this holder knows about, as a ranked directory answer."""
try:
return TransactiveService(settings).who_knows(query, as_holder=holder_id)
except ValueError as exc:
raise HTTPException(status_code=422, detail=str(exc)) from exc
@app.post("/api/who_knows", response_model=WhoKnowsReport)
def who_knows(
request: WhoKnowsRequest,
settings: Annotated[Settings, Depends(settings_dep)],
) -> WhoKnowsReport:
try:
return TransactiveService(settings).who_knows(
request.query, as_holder=request.as_holder, limit=request.limit
)
except ValueError as exc:
raise HTTPException(status_code=422, detail=str(exc)) from exc
@app.post("/api/ask", response_model=ReferralAnswer)
def ask(
request: AskRequest,
settings: Annotated[Settings, Depends(settings_dep)],
) -> ReferralAnswer:
try:
return TransactiveService(settings).ask(
request.question,
as_holder=request.as_holder,
max_hops=request.max_hops,
include_text=request.include_text,
)
except ValueError as exc:
raise HTTPException(status_code=422, detail=str(exc)) from exc
+6 -2
View File
@@ -16,6 +16,7 @@ from .benchmarks import (
)
from .cartridge import CartridgeService
from .cartridge_assets import CartridgeAssetService
from .cli_transactive import register as register_transactive_commands
from .config import get_settings
from .db import Database
from .embeddings import EmbeddingService
@@ -38,6 +39,9 @@ def json_dump(value: object) -> str:
return json.dumps(value, ensure_ascii=False, indent=2)
register_transactive_commands(app, lambda: get_settings())
@app.command()
def serve(host: str = "127.0.0.1", port: int = 8765) -> None:
uvicorn.run(create_app(), host=host, port=port)
@@ -202,7 +206,7 @@ def unmount(cartridge_id: str) -> None:
@app.command()
def append(
cartridge_id: str,
atom_id: list[str] = typer.Option(..., "--atom-id"), # noqa: B008 # typer reads option defaults itself
atom_id: list[str] = typer.Option(..., "--atom-id"),
) -> None:
"""Add memories to an existing cartridge; the bundle and its DNA follow."""
service = CartridgeService(get_settings())
@@ -352,7 +356,7 @@ def build_visual_training_data(
prd_path: Path = Path("PRD.md"),
output_path: Path | None = None,
cartridge_id: str | None = None,
image_path: list[Path] | None = typer.Option(None, "--image-path"), # noqa: B008 # typer reads option defaults itself
image_path: list[Path] | None = typer.Option(None, "--image-path"),
) -> None:
path = TrainingService(get_settings()).build_visual_dataset(
prd_path=prd_path,
+100
View File
@@ -0,0 +1,100 @@
"""CLI face for the directory: register holders, ask who knows, ask and get referred.
Registered onto the main Typer app so the command names stay flat (`ely-eye who-knows`), while
the transactive surface stays in one file.
"""
from __future__ import annotations
from collections.abc import Callable
import typer
from .config import Settings
from .db import Database
from .holder_store import HolderStore
from .schemas import utc_now
from .transactive import TransactiveService
from .transactive_schemas import Holder, HolderKind
def register(app: typer.Typer, settings: Callable[[], Settings]) -> None:
"""Add the holder and referral commands. The caller passes a settings thunk so one patch
in a test redirects every command, the same way the cartridge routes do."""
def store() -> HolderStore:
return HolderStore(Database(settings()))
def asker(as_holder: str) -> str:
"""An empty --as-holder means the local principal, the same identity `remember` writes as."""
return as_holder or settings().default_holder
@app.command("holders")
def holders() -> None:
"""List registered holders: the directory a referral walks."""
for entry in store().list_holders():
typer.echo(entry.model_dump_json(indent=2))
@app.command("holder-set")
def holder_set(
holder_id: str,
name: str = "",
kind: HolderKind = "person",
knows_about: list[str] = typer.Option(None, "--knows-about"),
contact: list[str] = typer.Option(None, "--contact", help="key=value, repeatable"),
) -> None:
"""Create or replace one holder entry."""
now = utc_now()
entry = Holder(
holder_id=holder_id,
name=name,
kind=kind,
knows_about=list(knows_about or []),
contact=_pairs(contact),
created_at=now,
updated_at=now,
)
typer.echo(store().upsert(entry).model_dump_json(indent=2))
@app.command("holder-rm")
def holder_rm(holder_id: str) -> None:
"""Delete a holder. Refuses while that holder still owns memories."""
store().delete(holder_id)
typer.echo(f'{{"deleted": "{holder_id}"}}')
@app.command("who-knows")
def who_knows(query: str, as_holder: str = "", limit: int = 8) -> None:
"""Rank the holders whose memories answer this query."""
report = TransactiveService(settings()).who_knows(
query, as_holder=asker(as_holder), limit=limit
)
typer.echo(report.model_dump_json(indent=2))
@app.command()
def ask(
question: str,
as_holder: str = "",
max_hops: int = 3,
include_text: bool = typer.Option(
False, "--include-text", help="show the memory text when the asker may already read it"
),
) -> None:
"""Ask the store, then whoever knows. An unanswered question says nobody knows."""
answer = TransactiveService(settings()).ask(
question,
as_holder=asker(as_holder),
max_hops=max_hops,
include_text=include_text,
)
typer.echo(answer.model_dump_json(indent=2))
def _pairs(values: list[str] | None) -> dict[str, str]:
"""Turn repeatable key=value options into a contact map, naming any entry that is not one."""
contact: dict[str, str] = {}
for value in values or []:
key, separator, detail = value.partition("=")
if not separator or not key.strip():
raise ValueError(f"--contact needs key=value, got {value!r}")
contact[key.strip()] = detail
return contact
+16
View File
@@ -188,12 +188,18 @@ class Database:
until: datetime | None = None,
min_trust: float | None = None,
cartridge_ids: list[str] | None = None,
as_holder: str | None = None,
) -> FtsPage:
"""BM25 over the FTS index joined to memories that are open right now.
Every filter is a `? IS NULL OR` predicate in one literal statement: the ranking never
depends on string-built SQL, and the planner sees one shape. A memory with no event time
cannot answer a time filter, so it drops out when one is asked for.
`as_holder` applies the sharing rule: a memory is readable by its owner, by anyone named
in its visible_to list, by everyone when it is public, and by anyone when nobody owns it.
A memory with no holder cannot be private to a holder that does not exist. Without an
asker the query runs over the whole library, which is what an operator debugging wants.
"""
with self.connect() as connection:
rows = connection.execute(
@@ -221,6 +227,13 @@ class Database:
WHERE ca.atom_id = f.atom_id
AND ca.cartridge_id IN (SELECT value FROM json_each(?))
))
AND (? IS NULL OR a.holder_id = ''
OR json_extract(a.visibility_json, '$.scope') = 'everyone'
OR a.holder_id = ?
OR EXISTS (
SELECT 1 FROM json_each(json_extract(a.visibility_json, '$.holders'))
WHERE value = ?
))
ORDER BY score
LIMIT ? OFFSET ?
""",
@@ -242,6 +255,9 @@ class Database:
min_trust,
None if cartridge_ids is None else to_json(cartridge_ids),
None if cartridge_ids is None else to_json(cartridge_ids),
as_holder,
as_holder,
as_holder,
limit + 1,
offset,
),
+173
View File
@@ -0,0 +1,173 @@
"""The holder registry and the directory query: who holds a memory, and who knows a topic.
A transactive memory system stores the directory, not the content: "I do not know where the
keys are, I know my wife knows, and I can find my wife." That requires a holder entity the
store can answer questions about, and it requires the answer to be a pointer with a reason,
never a guess.
"""
from __future__ import annotations
import sqlite3
from dataclasses import dataclass
from datetime import datetime
from typing import cast
from .db import Database
from .row_codec import from_json, to_json
from .schemas import utc_now
from .text_index import fts_query
from .transactive_schemas import Holder, HolderKind
@dataclass(frozen=True)
class WhoKnowsCount:
"""One holder's evidence for one query, straight from the index."""
holder_id: str
matched: int
top_score: float
latest_time: str | None
class HolderStore:
"""CRUD for registered holders plus the raw counts who_knows is built from."""
def __init__(self, database: Database) -> None:
self.database = database
def upsert(self, holder: Holder) -> Holder:
"""Insert or replace a registry entry, keeping the original creation time."""
now = utc_now().isoformat()
with self.database.connect() as connection:
existing = connection.execute(
"SELECT created_at FROM holders WHERE holder_id = ?", (holder.holder_id,)
).fetchone()
created_at = str(existing["created_at"]) if existing is not None else now
stored = holder.model_copy(update={"created_at": _moment(created_at), "updated_at": _moment(now)})
connection.execute(
"""
INSERT INTO holders (holder_id, name, kind, knows_about_json, contact_json, created_at, updated_at)
VALUES (?, ?, ?, ?, ?, ?, ?)
ON CONFLICT(holder_id) DO UPDATE SET
name = excluded.name,
kind = excluded.kind,
knows_about_json = excluded.knows_about_json,
contact_json = excluded.contact_json,
updated_at = excluded.updated_at
""",
(
stored.holder_id,
stored.name,
stored.kind,
to_json(stored.knows_about),
to_json(stored.contact),
stored.created_at.isoformat(),
stored.updated_at.isoformat(),
),
)
return stored
def get(self, holder_id: str) -> Holder | None:
with self.database.connect() as connection:
row = connection.execute("SELECT * FROM holders WHERE holder_id = ?", (holder_id,)).fetchone()
return None if row is None else _row_to_holder(row)
def list_holders(self) -> list[Holder]:
with self.database.connect() as connection:
rows = connection.execute("SELECT * FROM holders ORDER BY holder_id").fetchall()
return [_row_to_holder(row) for row in rows]
def delete(self, holder_id: str) -> None:
"""Remove a registry entry. Refuses while that holder still owns memories: forgetting a
phone number must not silently orphan the memories it points to."""
holding = self.atom_count(holder_id)
if holding:
raise ValueError(
f"holder {holder_id!r} still holds {holding} memories; forget them before deleting the holder"
)
with self.database.connect() as connection:
connection.execute("DELETE FROM holders WHERE holder_id = ?", (holder_id,))
def atom_count(self, holder_id: str) -> int:
with self.database.connect() as connection:
row = connection.execute(
"SELECT COUNT(*) AS n FROM v_atoms_current WHERE holder_id = ?", (holder_id,)
).fetchone()
return int(row["n"])
def known_holders(self) -> list[str]:
"""Every holder with at least one open memory, registered or not.
A person who has memories is a holder whether or not anybody registered them; the
registry adds declared topics and contact details, it does not grant existence.
"""
with self.database.connect() as connection:
rows = connection.execute(
"SELECT DISTINCT holder_id FROM v_atoms_current WHERE holder_id <> '' ORDER BY holder_id"
).fetchall()
return [str(row["holder_id"]) for row in rows]
def who_knows_counts(self, query: str, limit: int) -> list[WhoKnowsCount]:
"""Rank holders by how well their own memories answer the query.
The directory is deliberately unfiltered by visibility: the whole point of transactive
memory is knowing that someone knows, even when their memory is private to them.
`rank` is FTS5's own bm25 column, and it is the only way to aggregate the score: SQLite
refuses bm25() inside MAX(), and refuses it in a subquery or window over a match.
"""
with self.database.connect() as connection:
rows = connection.execute(
"""
SELECT a.holder_id,
COUNT(*) AS matched,
-MAX(rank) AS top_score,
MAX(COALESCE(a.event_time, a.ingested_at)) AS latest_time
FROM atom_fts AS f
JOIN v_atoms_current AS a ON a.atom_id = f.atom_id
WHERE atom_fts MATCH ?
AND a.holder_id <> ''
GROUP BY a.holder_id
ORDER BY top_score DESC, matched DESC
LIMIT ?
""",
(fts_query(query), limit),
).fetchall()
return [
WhoKnowsCount(
holder_id=str(row["holder_id"]),
matched=int(row["matched"]),
top_score=float(row["top_score"]),
latest_time=None if row["latest_time"] is None else str(row["latest_time"]),
)
for row in rows
]
def _moment(value: str) -> datetime:
parsed = datetime.fromisoformat(value)
if parsed.tzinfo is None:
raise ValueError(f"holder timestamps need an explicit offset, stored {value!r}")
return parsed
_HOLDER_KINDS: tuple[HolderKind, ...] = ("person", "agent", "service")
def _kind(value: str) -> HolderKind:
if value not in _HOLDER_KINDS:
raise ValueError(f"stored holder kind {value!r} is not one of person, agent, service")
return cast(HolderKind, value)
def _row_to_holder(row: sqlite3.Row) -> Holder:
return Holder(
holder_id=str(row["holder_id"]),
name=str(row["name"]),
kind=_kind(str(row["kind"])),
knows_about=from_json(row["knows_about_json"], []),
contact=from_json(row["contact_json"], {}),
created_at=_moment(str(row["created_at"])),
updated_at=_moment(str(row["updated_at"])),
)
+16 -1
View File
@@ -14,7 +14,7 @@ import sqlite3
from .text_index import index_text
SCHEMA_VERSION = 4
SCHEMA_VERSION = 5
ATOM_COLUMNS_V2 = """
atom_id TEXT PRIMARY KEY,
@@ -82,6 +82,21 @@ _BASE_TABLES = """
mounted_at TEXT NOT NULL
);
-- The directory of who may hold or answer a memory. A transactive system answers
-- "who knows this?" before it answers "what is this?", and it cannot do that without
-- an entity to point at. Contact details are free-form on purpose: the transport is
-- whoever wired this store together, not the schema.
CREATE TABLE IF NOT EXISTS holders (
holder_id TEXT PRIMARY KEY,
name TEXT NOT NULL DEFAULT '',
kind TEXT NOT NULL DEFAULT 'person'
CHECK (kind IN ('person', 'agent', 'service')),
knows_about_json TEXT NOT NULL DEFAULT '[]',
contact_json TEXT NOT NULL DEFAULT '{}',
created_at TEXT NOT NULL,
updated_at TEXT NOT NULL
);
CREATE TABLE IF NOT EXISTS cache_events (
id INTEGER PRIMARY KEY AUTOINCREMENT,
ts TEXT NOT NULL,
+3 -1
View File
@@ -157,10 +157,12 @@ class SearchFilters(BaseModel):
"""Who may answer, in what setting, and how fresh the memory has to be.
A memory with no event time cannot satisfy since/until, so it drops out when a time bound is
asked for rather than passing as unknown.
asked for rather than passing as unknown. `holder` restricts the answer to one holder's
memories; `as_holder` is the identity asking, and applies the sharing rule instead.
"""
holder: str | None = None
as_holder: str | None = None
scenario: str | None = None
source_id: str | None = None
modality: Modality | None = None
+13 -1
View File
@@ -42,7 +42,7 @@ def fts_query(query: str) -> str:
terms = _bare(query).split()
if not terms:
raise ValueError(f"search needs at least one searchable word, got {query!r}")
return " ".join(_phrase(term) if _is_cjk(term) else f"{_quote(term)}*" for term in terms)
return " ".join(_cjk_phrase(term) if _is_cjk(term) else f"{_quote(term)}*" for term in terms)
def _bare(text: str) -> str:
@@ -59,5 +59,17 @@ def _phrase(run: str) -> str:
return '"' + " ".join(run) + '"'
def _cjk_phrase(run: str) -> str:
"""A CJK run stays adjacent, and a long question also reaches the pairs inside it.
The whole run as one phrase is the precise reading, and it is what a two-character word
needs. A question is usually longer than the sentence that answers it: 钥匙在哪里 will not
appear inside 我们的钥匙放在玄关抽屉里. Offering every adjacent pair as an alternative keeps
adjacency — 钥 and 匙 still have to touch — without demanding equal length.
"""
phrases = list(dict.fromkeys([_phrase(run)] + [_phrase(run[start : start + 2]) for start in range(len(run) - 1)]))
return phrases[0] if len(phrases) == 1 else "(" + " OR ".join(phrases) + ")"
def _quote(term: str) -> str:
return '"' + term.replace('"', '""') + '"'
+223
View File
@@ -0,0 +1,223 @@
"""Ask the store who knows something, then ask them.
The transactive rule this service implements: not knowing is fine when the directory knows
where to send you. Three things follow, and each one is a failure mode if left implicit.
- The directory is public, the content is not. `who_knows` counts every holder's matching
memories, including memories the asker may not read, because "she knows" is the answer.
- A referral never invents an answer. If the asker may not read the holder's memories, the
result is a pointer plus contact details, and the text stays where it is.
- Referral must terminate. Every hop records whom it asked, and the chain stops at the hop
limit or when every candidate has been asked once.
"""
from __future__ import annotations
from datetime import datetime
from typing import Literal
from .config import Settings, get_settings
from .db import Database
from .holder_store import HolderStore, WhoKnowsCount
from .retrieval import RetrievalService
from .schemas import SearchFilters
from .text_index import index_text
from .transactive_schemas import (
ReferralAnswer,
WhoKnowsHit,
WhoKnowsReport,
)
NO_HOLDER_KNOWS: Literal["no_holder_knows"] = "no_holder_knows"
LIBRARY = "library"
WHO_KNOWS_LIMIT = 8
MAX_HOPS_LIMIT = 10
# Confidence is ordinal: it orders candidates, it does not estimate a probability. The weights
# say what the evidence is worth relative to itself, and every hit carries the raw numbers too.
CONFIDENCE_SATURATE = 3
WEIGHT_STRENGTH = 0.60
WEIGHT_COVERAGE = 0.25
WEIGHT_DECLARED = 0.15
class TransactiveService:
"""who_knows and ask, built on the holder registry and the same retrieval as search."""
def __init__(self, settings: Settings | None = None, database: Database | None = None) -> None:
self.settings = settings or get_settings()
self.db = database or Database(self.settings)
self.holders = HolderStore(self.db)
self.retrieval = RetrievalService(self.settings, self.db)
def who_knows(
self, query: str, *, as_holder: str | None = None, limit: int = WHO_KNOWS_LIMIT
) -> WhoKnowsReport:
"""Rank the holders whose own memories answer this query.
An empty result is the honest answer, not a fallback: nobody holds this, so the caller
should say so instead of paraphrasing something unrelated.
"""
if not query.strip():
raise ValueError("who_knows needs a query")
if limit <= 0:
raise ValueError(f"who_knows needs a positive limit, got {limit}")
counts = self.holders.who_knows_counts(query, limit)
if not counts:
return WhoKnowsReport(query=query, as_holder=as_holder, status=NO_HOLDER_KNOWS, hits=[])
registered = {holder.holder_id: holder for holder in self.holders.list_holders()}
best = max(count.top_score for count in counts)
return WhoKnowsReport(
query=query,
as_holder=as_holder,
status="found",
hits=[
self._hit(count, best, registered.get(count.holder_id), query)
for count in counts
],
)
def ask(
self,
question: str,
*,
as_holder: str | None = None,
max_hops: int = 3,
include_text: bool = False,
) -> ReferralAnswer:
"""Answer a question by asking the asker first, then the directory's best candidates.
`answered_by` is the asker when their own memory answered, `library` when an unowned
artifact did, and the holder whose memory answered otherwise. `referred` means somebody
knows but their memory is not readable by the asker.
`max_hops` counts the holders asked besides the asker: the asker's own memory is hop 0,
so max_hops=1 still asks the best candidate.
"""
if not question.strip():
raise ValueError("ask needs a question")
if not 1 <= max_hops <= MAX_HOPS_LIMIT:
raise ValueError(f"ask needs max_hops between 1 and {MAX_HOPS_LIMIT}, got {max_hops}")
own = self._search(question, as_holder=as_holder, holder=None)
if own:
return self._answered(
question, as_holder, asked=[as_holder or LIBRARY], hops=0, hits=own, include_text=include_text
)
asked: list[str] = [as_holder or LIBRARY]
for hit in self.who_knows(question, as_holder=as_holder).hits:
if hit.holder_id in asked:
continue
if len(asked) - 1 >= max_hops:
break
asked.append(hit.holder_id)
readable = self._search(question, as_holder=as_holder, holder=hit.holder_id)
if readable:
return self._answered(
question,
as_holder,
asked=asked,
hops=len(asked) - 1,
hits=readable,
include_text=include_text,
)
return self._pointer(question, as_holder, asked, hit)
return ReferralAnswer(
question=question,
as_holder=as_holder,
status=NO_HOLDER_KNOWS,
asked=asked,
hops=len(asked) - 1,
basis=["no holder has a memory that answers this"],
)
def _search(self, query: str, *, as_holder: str | None, holder: str | None):
result = self.retrieval.search(
query,
top_k=5,
filters=SearchFilters(holder=holder, as_holder=as_holder),
)
return result.hits
def _answered(self, question, as_holder, *, asked, hops, hits, include_text) -> ReferralAnswer:
"""The hits already passed the visibility clause in SQL, so they may be shown."""
top = hits[0]
owner = top.atom.holder_id or LIBRARY
return ReferralAnswer(
question=question,
as_holder=as_holder,
status="answered",
answered_by=owner,
asked=asked,
hops=hops,
atom_ids=[hit.atom.atom_id for hit in hits],
answer=top.atom.text if include_text else None,
basis=[
f"{len(hits)} memories readable by {as_holder or 'anyone'}",
f"best match scores {top.final_score:.2f}",
f"held by {owner}",
],
)
def _pointer(self, question, as_holder, asked, hit: WhoKnowsHit) -> ReferralAnswer:
"""The candidate knows but the asker cannot read it: return the contact, not the text."""
contact = hit.contact
return ReferralAnswer(
question=question,
as_holder=as_holder,
status="referred",
answered_by=hit.holder_id,
asked=asked,
hops=len(asked) - 1,
basis=[
f"{hit.holder_id} holds {hit.matched} matching memories the asker may not read",
f"best match scores {hit.top_score:.2f}",
f"reachable via {contact}" if contact else "no contact details registered",
],
)
def _hit(self, count: WhoKnowsCount, best: float, holder, query: str) -> WhoKnowsHit:
declared = _declared_topics(holder.knows_about if holder else [], query)
strength = count.top_score / best if best > 0 else 0.0
coverage = min(1.0, count.matched / CONFIDENCE_SATURATE)
confidence = (
WEIGHT_STRENGTH * strength + WEIGHT_COVERAGE * coverage + WEIGHT_DECLARED * (1.0 if declared else 0.0)
)
basis = [
f"holds {count.matched} matching memories",
f"best match scores {count.top_score:.2f}",
]
if count.latest_time:
basis.append(f"most recent memory is from {count.latest_time}")
if declared:
basis.append(f"declared topics: {', '.join(declared)}")
contact = dict(holder.contact) if holder else {}
if contact:
basis.append(f"reachable via {contact}")
return WhoKnowsHit(
holder_id=count.holder_id,
name=holder.name if holder else "",
confidence=round(confidence, 3),
matched=count.matched,
top_score=round(count.top_score, 3),
latest_time=None if count.latest_time is None else _as_moment(count.latest_time),
basis=basis,
contact=contact,
)
def _as_moment(value: str) -> datetime:
"""A stored timestamp that will not parse is a corrupt row, not a missing one."""
try:
return datetime.fromisoformat(value)
except ValueError as error:
raise ValueError(f"stored timestamp {value!r} is unreadable: {error}") from error
def _declared_topics(topics: list[str], query: str) -> list[str]:
"""Declared topics that share a searchable token with the query, using the index's own split."""
query_tokens = set(index_text(query).split())
if not query_tokens:
return []
return [topic for topic in topics if query_tokens & set(index_text(topic).split())]
+115
View File
@@ -0,0 +1,115 @@
"""Models for the directory side of memory: who exists, who knows what, and who answered.
Confidence here is ordinal, not calibrated: it orders candidates by the evidence in front of
us. Every model that carries a score also carries the raw evidence behind it, so a caller can
disagree with the number without re-running the query.
"""
from __future__ import annotations
from datetime import datetime
from typing import Literal
from pydantic import BaseModel, Field, field_validator
HolderKind = Literal["person", "agent", "service"]
"""The registry's own vocabulary; migrations.py enforces the same set in the database."""
class Holder(BaseModel):
"""A registered person, agent, or service that can hold or answer memories."""
holder_id: str
name: str = ""
kind: HolderKind = "person"
knows_about: list[str] = Field(default_factory=list)
contact: dict[str, str] = Field(default_factory=dict)
created_at: datetime
updated_at: datetime
@field_validator("holder_id")
@classmethod
def _holder_id_is_usable(cls, value: str) -> str:
if not value.strip():
raise ValueError("holder_id cannot be blank; a directory entry has to name somebody")
return value.strip()
class HolderUpsertRequest(BaseModel):
"""What a caller may set about a holder. The id comes from the path, never the body."""
name: str = ""
kind: HolderKind = "person"
knows_about: list[str] = Field(default_factory=list)
contact: dict[str, str] = Field(default_factory=dict)
class WhoKnowsHit(BaseModel):
"""One candidate holder, the evidence for them, and how to reach them."""
holder_id: str
name: str = ""
confidence: float
matched: int
top_score: float
latest_time: datetime | None = None
basis: list[str] = Field(default_factory=list)
contact: dict[str, str] = Field(default_factory=dict)
class WhoKnowsReport(BaseModel):
"""The directory's answer to a question. `no_holder_knows` is a result, not a failure."""
query: str
as_holder: str | None = None
status: Literal["found", "no_holder_knows"]
hits: list[WhoKnowsHit] = Field(default_factory=list)
class WhoKnowsRequest(BaseModel):
"""Ask the directory who holds an answer."""
query: str
as_holder: str | None = None
limit: int = Field(default=8, gt=0)
@field_validator("query")
@classmethod
def _query_is_usable(cls, value: str) -> str:
if not value.strip():
raise ValueError("who_knows needs a query")
return value.strip()
class AskRequest(BaseModel):
"""Ask a question on behalf of one holder, allowing a bounded chain of referrals."""
question: str
as_holder: str | None = None
max_hops: int = 3
include_text: bool = False
@field_validator("question")
@classmethod
def _question_is_usable(cls, value: str) -> str:
if not value.strip():
raise ValueError("ask needs a question")
return value.strip()
class ReferralAnswer(BaseModel):
"""Who answered, who was asked along the way, and what may be shown.
`referred` means the directory found someone who knows but the asker may not read their
memory: the answer is the pointer and the contact, never the text.
"""
question: str
as_holder: str | None = None
status: Literal["answered", "referred", "no_holder_knows"]
answered_by: str | None = None
asked: list[str] = Field(default_factory=list)
hops: int = 0
atom_ids: list[str] = Field(default_factory=list)
answer: str | None = None
basis: list[str] = Field(default_factory=list)
+249
View File
@@ -0,0 +1,249 @@
"""The transactive test: I do not know where the keys are, I know my wife knows, I find my wife."""
from __future__ import annotations
from pathlib import Path
from typing import Any
import pytest
from fastapi.testclient import TestClient
from pydantic import ValidationError
from typer.testing import CliRunner
import ely_eye.api as api_module
import ely_eye.cli as cli_module
from ely_eye.cli import app
from ely_eye.config import Settings
from ely_eye.db import Database
from ely_eye.ingestion import IngestionService
from ely_eye.memory import MemoryService
from ely_eye.migrations import SCHEMA_VERSION
from ely_eye.schemas import MemoryOrigin, Visibility, utc_now
from ely_eye.transactive import LIBRARY, NO_HOLDER_KNOWS, TransactiveService
from ely_eye.transactive_schemas import Holder
HOME_KEYS = "我们的钥匙放在玄关抽屉里"
WORK_KEYS = "备用钥匙在办公室第二个抽屉里"
QUERY = "钥匙"
class Store:
"""One SQLite file, the writer and the directory that read it."""
def __init__(self, settings: Settings) -> None:
self.settings = settings
self.memory = MemoryService(settings=settings)
self.db = self.memory.db
self.transactive = TransactiveService(settings=settings, database=self.db)
def wife_remembers(self, text: str, visibility: Visibility | None = None) -> str:
atom = self.memory.remember(
text, holder_id="wife", origin=MemoryOrigin.self_, scenario="home", visibility=visibility
)
return atom.atom_id
def husband_remembers(self, text: str) -> str:
atom = self.memory.remember(text, holder_id="husband", origin=MemoryOrigin.self_, scenario="home")
return atom.atom_id
def register(self, holder_id: str, **fields: Any) -> Holder:
moment = utc_now()
return self.transactive.holders.upsert(
Holder(holder_id=holder_id, created_at=moment, updated_at=moment, **fields)
)
@pytest.fixture
def store(tmp_path: Path) -> Store:
return Store(Settings(home=tmp_path / "home", default_holder="husband"))
def test_the_directory_names_the_holder_and_says_why(store: Store) -> None:
store.wife_remembers(HOME_KEYS)
report = store.transactive.who_knows(QUERY, as_holder="husband")
assert report.status == "found"
assert report.hits[0].holder_id == "wife"
assert report.hits[0].confidence > 0.0
assert any("matching memories" in line for line in report.hits[0].basis)
def test_the_directory_counts_a_memory_the_asker_cannot_read(store: Store) -> None:
"""The whole point: knowing that she knows does not require reading her memory."""
store.wife_remembers(HOME_KEYS)
assert store.transactive.who_knows(QUERY, as_holder="husband").hits[0].matched == 1
assert store.transactive._search(QUERY, as_holder="husband", holder="wife") == []
def test_a_private_memory_is_referred_and_not_leaked(store: Store) -> None:
store.register("wife", name="小美", contact={"channel": "signal", "address": "+86-555"})
store.wife_remembers(HOME_KEYS)
answer = store.transactive.ask("钥匙在哪里", as_holder="husband")
assert answer.status == "referred"
assert answer.answered_by == "wife"
assert answer.hops == 1
assert answer.asked == ["husband", "wife"]
assert answer.answer is None
assert answer.atom_ids == []
assert any("may not read" in line for line in answer.basis)
assert any("signal" in line for line in answer.basis)
def test_the_askers_own_memory_answers_without_a_referral(store: Store) -> None:
store.wife_remembers(HOME_KEYS)
store.husband_remembers("我的工牌挂在显示器边上")
answer = store.transactive.ask("工牌", as_holder="husband")
assert answer.status == "answered"
assert answer.answered_by == "husband"
assert answer.hops == 0
assert answer.asked == ["husband"]
assert answer.answer is None
def test_text_shown_only_when_the_asker_may_already_read_it(store: Store) -> None:
store.husband_remembers("我的工牌挂在显示器边上")
with_text = store.transactive.ask("工牌", as_holder="husband", include_text=True)
assert with_text.answer == "我的工牌挂在显示器边上"
store.register("wife")
store.wife_remembers(HOME_KEYS)
referred = store.transactive.ask("钥匙在哪里", as_holder="husband", include_text=True)
assert referred.status == "referred"
assert referred.answer is None
def test_a_shared_memory_is_readable_without_a_referral(store: Store) -> None:
"""A memory the asker may already read answers at hop 0: no referral, the text is shown."""
store.register("wife")
store.wife_remembers(HOME_KEYS, visibility=Visibility(scope="visible_to", holders=["husband"]))
answer = store.transactive.ask("钥匙在哪里", as_holder="husband", include_text=True)
assert answer.status == "answered"
assert answer.answered_by == "wife"
assert answer.hops == 0
assert answer.answer == HOME_KEYS
def test_an_unowned_artifact_answers_as_the_library(store: Store) -> None:
notebook = store.settings.home / "notebook.txt"
notebook.write_text("the garage door code is written on the fuse box", encoding="utf-8")
IngestionService(store.settings).ingest_path(notebook)
answer = store.transactive.ask("garage door code", as_holder="wife", include_text=True)
assert answer.status == "answered"
assert answer.answered_by == LIBRARY
assert "fuse box" in (answer.answer or "")
def test_nobody_knows_is_a_result_and_not_a_fabrication(store: Store) -> None:
store.wife_remembers(HOME_KEYS)
report = store.transactive.who_knows("潜水镜", as_holder="husband")
assert report.status == NO_HOLDER_KNOWS
assert report.hits == []
answer = store.transactive.ask("潜水镜在哪里", as_holder="husband")
assert answer.status == NO_HOLDER_KNOWS
assert answer.answered_by is None
assert answer.answer is None
assert answer.basis == ["no holder has a memory that answers this"]
def test_the_referral_walks_once_and_never_asks_the_same_holder_twice(store: Store) -> None:
store.register("wife")
store.register("daughter")
store.wife_remembers(HOME_KEYS)
store.memory.remember(WORK_KEYS, holder_id="daughter", origin=MemoryOrigin.self_, scenario="work")
answer = store.transactive.ask("钥匙在哪里", as_holder="husband", max_hops=1)
assert answer.status in ("referred", "answered")
assert len(answer.asked) == 2
assert len(set(answer.asked)) == len(answer.asked)
assert answer.asked[0] == "husband"
def test_the_hop_limit_is_enforced(store: Store) -> None:
for bad in (0, 11):
with pytest.raises(ValueError, match="max_hops between 1 and 10"):
store.transactive.ask(QUERY, as_holder="husband", max_hops=bad)
def test_declared_topics_raise_the_confidence_and_say_so(store: Store) -> None:
store.wife_remembers(HOME_KEYS)
before = store.transactive.who_knows(QUERY, as_holder="husband").hits[0]
store.register("wife", knows_about=["家里的钥匙"])
after = store.transactive.who_knows(QUERY, as_holder="husband").hits[0]
assert after.confidence > before.confidence
assert any("declared topics" in line for line in after.basis)
def test_the_registry_refuses_to_delete_a_holder_who_still_holds(store: Store) -> None:
store.register("wife")
atom_id = store.wife_remembers(HOME_KEYS)
with pytest.raises(ValueError, match="still holds 1 memories"):
store.transactive.holders.delete("wife")
store.memory.forget(atom_id=atom_id)
store.transactive.holders.delete("wife")
assert store.transactive.holders.list_holders() == []
def test_an_unknown_holder_kind_is_refused_at_the_door() -> None:
with pytest.raises(ValidationError, match="'person', 'agent' or 'service'"):
Holder.model_validate({"holder_id": "ghost", "kind": "ghost", "created_at": utc_now(), "updated_at": utc_now()})
def test_a_blank_holder_id_is_refused() -> None:
with pytest.raises(ValidationError, match="directory entry has to name somebody"):
Holder.model_validate({"holder_id": " ", "created_at": utc_now(), "updated_at": utc_now()})
def test_a_store_without_holders_gains_the_table_on_reopen(tmp_path: Path) -> None:
settings = Settings(home=tmp_path / "home", default_holder="husband")
database = Database(settings)
with database.connect() as connection:
connection.execute("DROP TABLE holders")
connection.execute("PRAGMA user_version = 4")
reopened = Database(settings)
with reopened.connect() as connection:
assert connection.execute("PRAGMA user_version").fetchone()[0] == SCHEMA_VERSION
assert connection.execute("SELECT COUNT(*) AS n FROM holders").fetchone()["n"] == 0
def test_the_http_surface_serves_the_directory(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
settings = Settings(home=tmp_path / "home", default_holder="husband")
monkeypatch.setattr(api_module, "get_settings", lambda: settings)
client = TestClient(api_module.create_app())
store = Store(settings)
store.wife_remembers(HOME_KEYS)
created = client.post(
"/api/holders/wife",
json={"name": "小美", "knows_about": ["家里的钥匙"], "contact": {"channel": "signal"}},
)
assert created.status_code == 200
assert created.json()["holder_id"] == "wife"
assert client.get("/api/holders").json()[0]["name"] == "小美"
assert client.get("/api/holders/nobody").status_code == 404
knows = client.get("/api/holders/wife/knows", params={"query": QUERY}).json()
assert knows["status"] == "found" and knows["hits"][0]["holder_id"] == "wife"
asked = client.post("/api/ask", json={"question": "钥匙在哪里", "as_holder": "husband"}).json()
assert asked["status"] == "referred" and asked["answered_by"] == "wife"
who = client.post("/api/who_knows", json={"query": "钥匙", "as_holder": "husband"}).json()
assert who["hits"][0]["basis"]
blocked = client.delete("/api/holders/wife")
assert blocked.status_code == 409 and "still holds" in blocked.json()["detail"]
def test_the_cli_asks_and_reports_the_referral(tmp_path: Path, monkeypatch: pytest.MonkeyPatch) -> None:
settings = Settings(home=tmp_path / "home", default_holder="husband")
monkeypatch.setattr(cli_module, "get_settings", lambda: settings)
runner = CliRunner()
Store(settings).wife_remembers(HOME_KEYS)
assert runner.invoke(app, ["holder-set", "wife", "--name", "小美", "--contact", "channel=signal"]).exit_code == 0
who = runner.invoke(app, ["who-knows", QUERY])
assert who.exit_code == 0 and '"holder_id": "wife"' in who.stdout
asked = runner.invoke(app, ["ask", "钥匙在哪里"])
assert asked.exit_code == 0 and '"status": "referred"' in asked.stdout
assert runner.invoke(app, ["holders"]).stdout.count("wife") >= 1
assert runner.invoke(app, ["holder-set", "wife", "--contact", "not-a-pair"]).exit_code != 0
+40
View File
@@ -207,3 +207,43 @@ type ProofSuiteDetail struct {
Path string `json:"path"`
HashHop HashHopReport `json:"hashhop"`
}
type Holder struct {
HolderID string `json:"holder_id"`
Name string `json:"name"`
Kind string `json:"kind"`
KnowsAbout []string `json:"knows_about"`
Contact map[string]string `json:"contact"`
CreatedAt string `json:"created_at"`
UpdatedAt string `json:"updated_at"`
}
type WhoKnowsHit struct {
HolderID string `json:"holder_id"`
Name string `json:"name"`
Confidence float64 `json:"confidence"`
Matched int64 `json:"matched"`
TopScore float64 `json:"top_score"`
LatestTime string `json:"latest_time,omitempty"`
Basis []string `json:"basis"`
Contact map[string]string `json:"contact"`
}
type WhoKnowsReport struct {
Query string `json:"query"`
AsHolder string `json:"as_holder,omitempty"`
Status string `json:"status"`
Hits []WhoKnowsHit `json:"hits"`
}
type ReferralAnswer struct {
Question string `json:"question"`
AsHolder string `json:"as_holder,omitempty"`
Status string `json:"status"`
AnsweredBy string `json:"answered_by,omitempty"`
Asked []string `json:"asked"`
Hops int64 `json:"hops"`
AtomIDs []string `json:"atom_ids"`
Answer string `json:"answer,omitempty"`
Basis []string `json:"basis"`
}
@@ -0,0 +1,102 @@
package mcpserver
import (
"context"
"errors"
"github.com/ZacharyZhang-NY/ely-eye/ely-eye-mcp/internal/domain"
"github.com/modelcontextprotocol/go-sdk/mcp"
)
type WhoKnowsInput struct {
Query string `json:"query" jsonschema:"what the asker wants to know, in the asker's own words"`
AsHolder string `json:"as_holder,omitempty" jsonschema:"who is asking; decides which memories may be shown"`
Limit int `json:"limit,omitempty" jsonschema:"how many candidates to return, default 8"`
}
type AskInput struct {
Question string `json:"question" jsonschema:"the question to answer from memory"`
AsHolder string `json:"as_holder,omitempty" jsonschema:"who is asking; their own memory is checked first"`
MaxHops int `json:"max_hops,omitempty" jsonschema:"how many holders besides the asker may be asked, default 3"`
IncludeText bool `json:"include_text,omitempty" jsonschema:"show the memory text when the asker may already read it"`
}
// registerDirectoryTools exposes the transactive surface: who knows, and ask-and-get-referred.
func (s *Server) registerDirectoryTools(server *mcp.Server) {
mcp.AddTool(server, &mcp.Tool{
Title: "Who knows",
Name: "who_knows",
Description: "Rank the holders whose memories answer a query, with the evidence and how to reach them. The directory counts memories the asker cannot read; that is the point. no_holder_knows is a result, not a failure.",
}, s.whoKnows)
mcp.AddTool(server, &mcp.Tool{
Title: "Ask and get referred",
Name: "ask",
Description: "Ask the asker's own memory, then the directory. answered means somebody's memory matched; referred means the best holder's memory is private, so the answer is the pointer and contact, never the text.",
}, s.askMemory)
mcp.AddTool(server, &mcp.Tool{
Title: "List holders",
Name: "list_holders",
Description: "List registered holders: the people, agents, and services a referral can walk to.",
}, s.listHolders)
}
type whoKnowsOutput struct {
Report domain.WhoKnowsReport `json:"report" jsonschema:"the ranked candidates and their evidence"`
}
type askOutput struct {
Answer domain.ReferralAnswer `json:"answer" jsonschema:"who answered, who was asked, and what may be shown"`
}
type holdersOutput struct {
Holders []domain.Holder `json:"holders" jsonschema:"the registered holders"`
}
func (s *Server) whoKnows(ctx context.Context, _ *mcp.CallToolRequest, in WhoKnowsInput) (*mcp.CallToolResult, whoKnowsOutput, error) {
if in.Query == "" {
return nil, whoKnowsOutput{}, errors.New("query is required")
}
var report domain.WhoKnowsReport
body := map[string]any{"query": in.Query}
if in.AsHolder != "" {
body["as_holder"] = in.AsHolder
}
if in.Limit > 0 {
body["limit"] = in.Limit
}
if err := s.api.Post(ctx, "/api/who_knows", body, &report); err != nil {
return nil, whoKnowsOutput{}, err
}
return nil, whoKnowsOutput{Report: report}, nil
}
func (s *Server) askMemory(ctx context.Context, _ *mcp.CallToolRequest, in AskInput) (*mcp.CallToolResult, askOutput, error) {
if in.Question == "" {
return nil, askOutput{}, errors.New("question is required")
}
var answer domain.ReferralAnswer
body := map[string]any{"question": in.Question}
if in.AsHolder != "" {
body["as_holder"] = in.AsHolder
}
if in.MaxHops > 0 {
body["max_hops"] = in.MaxHops
}
if in.IncludeText {
body["include_text"] = true
}
if err := s.api.Post(ctx, "/api/ask", body, &answer); err != nil {
return nil, askOutput{}, err
}
return nil, askOutput{Answer: answer}, nil
}
func (s *Server) listHolders(ctx context.Context, _ *mcp.CallToolRequest, _ struct{}) (*mcp.CallToolResult, holdersOutput, error) {
var holders []domain.Holder
if err := s.api.Get(ctx, "/api/holders", &holders); err != nil {
return nil, holdersOutput{}, err
}
return nil, holdersOutput{Holders: holders}, nil
}
@@ -0,0 +1,133 @@
package mcpserver
import (
"context"
"encoding/json"
"io"
"net/http"
"net/http/httptest"
"strings"
"testing"
"github.com/ZacharyZhang-NY/ely-eye/ely-eye-mcp/internal/apiclient"
)
func directoryServer(t *testing.T, body string, seen *[]string) *httptest.Server {
t.Helper()
return httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
payload, _ := io.ReadAll(r.Body)
*seen = append(*seen, r.Method+" "+r.URL.Path+" "+string(payload))
w.WriteHeader(http.StatusOK)
_, _ = w.Write([]byte(body))
}))
}
func TestWhoKnowsNamesTheHolderAndKeepsTheEvidence(t *testing.T) {
var seen []string
server := directoryServer(t, `{"query":"钥匙","as_holder":"husband","status":"found","hits":[{
"holder_id":"wife","name":"小美","confidence":0.91,"matched":2,"top_score":0.02,
"basis":["wife holds 2 matching memories the asker may not read"],"contact":{"channel":"signal"}}]}`, &seen)
defer server.Close()
_, out, err := (&Server{api: apiclient.New(server.URL)}).whoKnows(
context.Background(), nil, WhoKnowsInput{Query: "钥匙", AsHolder: "husband"})
if err != nil {
t.Fatalf("who_knows failed: %v", err)
}
if out.Report.Status != "found" || len(out.Report.Hits) != 1 {
t.Fatalf("report = %+v, want one found candidate", out.Report)
}
hit := out.Report.Hits[0]
if hit.HolderID != "wife" || hit.Confidence != 0.91 || hit.Contact["channel"] != "signal" {
t.Errorf("hit = %+v, want the holder, her confidence, and how to reach her", hit)
}
if len(seen) != 1 || !strings.Contains(seen[0], "POST /api/who_knows") || !strings.Contains(seen[0], `"query":"钥匙"`) {
t.Fatalf("requests = %v, want one POST carrying the query", seen)
}
}
func TestWhoKnowsRefusesAnEmptyQueryBeforeTheNetwork(t *testing.T) {
var seen []string
server := directoryServer(t, `{}`, &seen)
defer server.Close()
if _, _, err := (&Server{api: apiclient.New(server.URL)}).whoKnows(context.Background(), nil, WhoKnowsInput{}); err == nil {
t.Fatal("an empty query must not reach the directory")
}
if len(seen) != 0 {
t.Errorf("requests = %v, want none", seen)
}
}
func TestAskReturnsThePointerAndNotThePrivateText(t *testing.T) {
var seen []string
server := directoryServer(t, `{"question":"钥匙在哪里","as_holder":"husband","status":"referred",
"answered_by":"wife","asked":["husband","wife"],"hops":1,"atom_ids":[],"basis":["wife holds 1 matching memories the asker may not read"]}`, &seen)
defer server.Close()
_, out, err := (&Server{api: apiclient.New(server.URL)}).askMemory(
context.Background(), nil, AskInput{Question: "钥匙在哪里", AsHolder: "husband", MaxHops: 2})
if err != nil {
t.Fatalf("ask failed: %v", err)
}
if out.Answer.Status != "referred" || out.Answer.Answer != "" {
t.Errorf("answer = %+v, want a referral with no text", out.Answer)
}
if out.Answer.AnsweredBy != "wife" || out.Answer.Hops != 1 {
t.Errorf("answer = %+v, want wife at hop 1", out.Answer)
}
if !strings.Contains(seen[0], `"max_hops":2`) || !strings.Contains(seen[0], `"as_holder":"husband"`) {
t.Errorf("request = %s, want the hop budget and the asker", seen[0])
}
}
func TestAskWithoutAnAskerOmitsTheIdentity(t *testing.T) {
var seen []string
server := directoryServer(t, `{"question":"garage code","status":"no_holder_knows","asked":[],"hops":0}`, &seen)
defer server.Close()
if _, _, err := (&Server{api: apiclient.New(server.URL)}).askMemory(context.Background(), nil, AskInput{Question: "garage code"}); err != nil {
t.Fatalf("ask failed: %v", err)
}
var body map[string]any
parts := strings.SplitN(seen[0], " ", 3)
if err := json.Unmarshal([]byte(parts[2]), &body); err != nil {
t.Fatalf("request body is not JSON: %s", seen[0])
}
if _, present := body["as_holder"]; present {
t.Errorf("body = %v, want as_holder omitted so the server applies its own default", body)
}
if _, present := body["max_hops"]; present {
t.Errorf("body = %v, want max_hops omitted; a Go zero is not a hop budget", body)
}
}
func TestAskRefusesAnEmptyQuestion(t *testing.T) {
var seen []string
server := directoryServer(t, `{}`, &seen)
defer server.Close()
if _, _, err := (&Server{api: apiclient.New(server.URL)}).askMemory(context.Background(), nil, AskInput{}); err == nil {
t.Fatal("an empty question must not be asked")
}
if len(seen) != 0 {
t.Errorf("requests = %v, want none", seen)
}
}
func TestListHoldersReadsTheRegistry(t *testing.T) {
var seen []string
server := directoryServer(t, `[{"holder_id":"wife","name":"小美","kind":"person","knows_about":["家里的钥匙"],"contact":{"channel":"signal"},"created_at":"2026-05-01T00:00:00+00:00","updated_at":"2026-05-01T00:00:00+00:00"}]`, &seen)
defer server.Close()
_, out, err := (&Server{api: apiclient.New(server.URL)}).listHolders(context.Background(), nil, struct{}{})
if err != nil {
t.Fatalf("list_holders failed: %v", err)
}
if len(out.Holders) != 1 || out.Holders[0].KnowsAbout[0] != "家里的钥匙" {
t.Fatalf("holders = %+v, want the declared topics", out.Holders)
}
if len(seen) != 1 || !strings.HasPrefix(seen[0], "GET /api/holders") {
t.Errorf("requests = %v, want one GET", seen)
}
}
+1
View File
@@ -37,6 +37,7 @@ func Run(ctx context.Context, cfg config.Config, version string) error {
})
server.registerTools(mcpServer)
server.registerWriteTools(mcpServer)
server.registerDirectoryTools(mcpServer)
server.registerMountTools(mcpServer)
server.registerResources(mcpServer)
if err := mcpServer.Run(ctx, &mcp.StdioTransport{}); err != nil {
+61 -3
View File
@@ -25,8 +25,10 @@ printf '{"metadata": {}, "weight_map": {"model-00001.safetensors": "0"}}' > "$mo
export ELY_EYE_MODEL_DIR="$model_dir"
api=""
wife_api=""
cleanup() {
if [ -n "$api" ]; then kill "$api" 2>/dev/null || true; fi
if [ -n "$wife_api" ]; then kill "$wife_api" 2>/dev/null || true; fi
rm -rf "$home"
}
trap cleanup EXIT
@@ -47,6 +49,27 @@ fi
(cd "$root/ely-eye-mcp" && go build -o "$home/ely-eye-mcp" ./cmd/ely-eye-mcp)
# A second principal on the same SQLite file. The server, not the client, decides who a memory
# belongs to, so the only way to write as the wife is to ask a server whose principal is the wife.
wife_port=$((port + 1))
ELY_EYE_DEFAULT_HOLDER=wife "$python_bin" -m ely_eye.cli serve --port "$wife_port" >"$home/wife-api.log" 2>&1 &
wife_api=$!
ready=0
for _ in $(seq 1 80); do
if curl -sf "http://127.0.0.1:${wife_port}/api/status" >/dev/null 2>&1; then ready=1; break; fi
sleep 0.5
done
if [ "$ready" -ne 1 ]; then
echo "FAIL: the wife's Ely-Eye API did not start"
cat "$home/wife-api.log"
exit 1
fi
secret="护照在书房第二个抽屉里"
curl -sf -X POST "http://127.0.0.1:${wife_port}/api/memory" -H 'content-type: application/json' \
-d "{\"text\":\"${secret}\",\"scenario\":\"home\"}" >/dev/null
curl -sf -X POST "http://127.0.0.1:${wife_port}/api/holders/wife" -H 'content-type: application/json' \
-d '{"name":"小美","kind":"person","knows_about":["护照"],"contact":{"channel":"signal","address":"+86-555"}}' >/dev/null
printf 'the pantry keeps flour\n' > "$home/pantry.txt"
uploaded="$(curl -sf -X POST "http://127.0.0.1:${port}/api/ingest/upload?cartridge_name=home" \
-F "files=@${home}/pantry.txt;type=text/plain")"
@@ -56,7 +79,7 @@ if [ -z "$cartridge_id" ] || [ "$cartridge_id" = "None" ]; then
exit 1
fi
"$python_bin" - "$home/ely-eye-mcp" "$cartridge_id" "$model_dir" <<'PY'
"$python_bin" - "$home/ely-eye-mcp" "$cartridge_id" "$model_dir" "$secret" <<'PY'
import json
import pathlib
import subprocess
@@ -65,6 +88,7 @@ import sys
binary = sys.argv[1]
cartridge_id = sys.argv[2]
model_dir = pathlib.Path(sys.argv[3])
secret = sys.argv[4]
proc = subprocess.Popen(
[binary, "server"],
stdin=subprocess.PIPE,
@@ -108,7 +132,8 @@ send({"jsonrpc": "2.0", "method": "notifications/initialized"})
names = [item["name"] for item in call(2, "tools/list", {})["tools"]]
for required in ("remember", "supersede_memory", "forget_memory", "mounted_cartridges",
"mount_cartridge", "switch_cartridge", "unmount_cartridge", "append_to_cartridge"):
"mount_cartridge", "switch_cartridge", "unmount_cartridge", "append_to_cartridge",
"who_knows", "ask", "list_holders"):
if required not in names:
raise SystemExit(f"FAIL: {required} is not registered; tools are {names}")
@@ -182,5 +207,38 @@ refused = call(3, "tools/call", {"name": "mount_cartridge", "arguments": {"cartr
if not refused.get("isError") or "rope" not in refused["content"][0]["text"]:
raise SystemExit(f"FAIL: a cartridge built for RoPE factor 4.0 mounted on a factor 2.0 machine: {refused}")
print("PASS: remember, supersede, search, forget and cartridge mounting work through MCP")
# The transactive proof. The asker is this machine's principal; the wife's memory is private to
# her, so the directory may say she knows without showing what she knows.
holders = tool("list_holders", {})["holders"]
if not any(item["holder_id"] == "wife" for item in holders):
raise SystemExit(f"FAIL: the registry does not list the wife: {holders}")
directory = tool("who_knows", {"query": "护照", "as_holder": "local"})["report"]
if directory["status"] != "found" or not directory["hits"]:
raise SystemExit(f"FAIL: the directory found nobody who knows: {directory}")
top = directory["hits"][0]
if top["holder_id"] != "wife" or top["contact"].get("channel") != "signal":
raise SystemExit(f"FAIL: the directory did not name the wife and how to reach her: {top}")
if not any("declared topics" in line for line in top["basis"]):
raise SystemExit(f"FAIL: her declared topic did not raise the confidence: {top['basis']}")
referred = tool("ask", {"question": "护照在哪里", "as_holder": "local", "include_text": True})["answer"]
if referred["status"] != "referred" or referred["answered_by"] != "wife":
raise SystemExit(f"FAIL: the question was not referred to the wife: {referred}")
if referred.get("answer"):
raise SystemExit(f"FAIL: a private memory leaked into the referral: {referred['answer']}")
if secret in json.dumps(referred, ensure_ascii=False):
raise SystemExit(f"FAIL: the referral carried the private text: {referred}")
if "wife" not in referred["asked"]:
raise SystemExit(f"FAIL: the referral never asked the wife: {referred}")
mine = tool("ask", {"question": "pantry flour", "as_holder": "local", "include_text": True})["answer"]
if mine["status"] != "answered" or not mine["atom_ids"]:
raise SystemExit(f"FAIL: the asker's own memory did not answer: {mine}")
nobody = tool("ask", {"question": "潜水镜", "as_holder": "local"})["answer"]
if nobody["status"] != "no_holder_knows" or nobody.get("answer"):
raise SystemExit(f"FAIL: nobody knowing was reported as an answer: {nobody}")
print("PASS: remember, supersede, search, forget, cartridge mounting and the transactive referral work through MCP")
PY