kyle/boxmaker
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Propose P21 (memory writes without approval, as in Hermes); revise the M5a spec to match

Browse Source 
Writes from sessions that have read untrusted content go to the
archive, labelled untrusted, and never reach the prompt. Every write is
checked, announced in its thread, journalled and can be undone.
Session-scoped approvals for other tools become an open decision.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
This commit is contained in:
kyle
2026-09-24 13:41:02 -07:00
co-authored by Claude Opus 5.5
parent 817fb9d47f
commit 29f9069a33
2 changed files with 156 additions and 70 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/specs/2026-09-24-m5a-local-memory.md
+151 -69
View File
@@ -1,8 +1,9 @@
# M5a design: local memory
Status: draft for the owner's review, 2026-09-24. It rests on the brief as changed by P17 to P20
(applied 2026-09-24, `docs/decisions.md`): memory is three files read into the baseline, edited
through a `memory` tool at `ask`; a new role, `memoryd`, owns the files and a full-text index; and
Status: draft for the owner's review, 2026-09-24, revised the same day: memory writes need no
approval (P21, proposed). It rests on the brief as changed by P17 to P20 (applied 2026-09-24,
`docs/decisions.md`) and P21: memory is three files read into the baseline, edited by the agent
through a `memory` tool without questions, as Hermes does; a new role, `memoryd`, owns the files and a full-text index; and
M5 is split into M5a (this), M5b (Honcho) and M5c (the scheduler). Owner's decisions of
2026-09-24, beyond those: `soul.md` starts from a generic file shipped in the repository, and
`memory.md` and `user.md` start empty; and compaction (new epochs) comes with the scheduler in
@@ -25,20 +26,24 @@ Facts this rests on, checked on 2026-09-24:
1. A fact from one session is recalled in another.
2. An entry moved out of `memory.md` by `consolidate` is found by `recall`.
3. A memory edit reaches the prompt only in the next session, and only after the owner approved
it; a session already running keeps the bytes it started with.
4. The baseline with the starter `soul.md` stays at 3,000 tokens or less, measured on the device.
5. A hit from a session that had seen untrusted content comes back marked untrusted.
3. A memory write reaches the prompt only in the next session; a session already running keeps
the bytes it started with. No write asks the owner anything.
4. A write from a session that has seen untrusted content lands in the archive, labelled
untrusted, and never in the prompt.
5. The baseline with the starter `soul.md` stays at 3,000 tokens or less, measured on the device.
6. A hit from a session that had seen untrusted content comes back marked untrusted.
7. The owner can see every write where it happened, and undo the latest one of a file.
## 2. Components
| Piece | Change |
|---|---|
| `memoryd` (new crate) | Owns `memory/` and `index/`. Serves the memory block to `loopd`, runs `recall` and `memory` calls for `brokerd`, indexes session logs, rebuilds the index for `bxctl`. No network. |
| `proto` | Five new message kinds (section 7); the `memory` tool's argument structs. |
| `loopd` | The baseline takes the memory block from `memoryd` instead of reading `memory/core.md`. `recall` and `memory` are core tools. After each finished turn it tells `memoryd` the session has new records. |
| `brokerd` | `recall` and `memory` are tools it decides like any other; a decided call goes to `memoryd` instead of a container. A result's label can only be made stricter by the runtime. |
| `bxctl` | `bxctl reindex`. |
| `proto` | Eight new message kinds (section 7); the `memory` tool's argument structs; `TurnEvent::MemoryWritten`. |
| `loopd` | The baseline takes the memory block from `memoryd` instead of reading `memory/core.md`. `recall` and `memory` are core tools. After each finished turn it tells `memoryd` the session has new records. A successful write is announced as a turn event. |
| `brokerd` | `recall` and `memory` are decided under built-in grants (both `auto`); a decided call goes to `memoryd` with the session's state, instead of a container. A result's label can only be made stricter by the runtime. |
| `gatewayd` | Posts a `MemoryWritten` event as one line in the thread. |
| `bxctl` | `bxctl reindex`; `bxctl memory log`, `undo`, and, when review is on, `pending`, `accept`, `reject`. |
| `config/soul.md` (new) | The generic starter persona. |
`memoryd` depends only on `proto` (and `rusqlite`), like every role.
@@ -51,10 +56,11 @@ any of them with an editor; `soul.md` only ever that way.
| File | Written by | Limit | Starts as |
|---|---|---|---|
| `soul.md` | the owner | 4,000 characters | a copy of the starter (`[memory] soul_starter`, default `soul.md` next to the config file) |
| `memory.md` | `memoryd`, applying approved edits | 2,200 characters | empty |
| `user.md` | `memoryd`, applying approved edits | 1,375 characters | empty |
| `archive.md` | `memoryd`, appending moved entries | none | empty |
| `memory.md` | `memoryd`, applying the agent's writes | 2,200 characters | empty |
| `user.md` | `memoryd`, applying the agent's writes | 1,375 characters | empty |
| `archive.md` | `memoryd`, appending moved and dropped entries, and writes from untrusted sessions | none | empty |
| `edits/YYYY-MM.jsonl` | `memoryd`, appending one line per applied edit | none | absent |
| `pending.jsonl` | `memoryd`, when review is on (section 5) | none | absent |
| `recalls/YYYY-MM.jsonl` | `memoryd`, appending one line per `recall` | none | absent |
At startup `memoryd` creates what is missing and never overwrites anything. Characters are
@@ -64,15 +70,23 @@ counted with `chars()`. Limits are on the file's text.
(as Hermes does). An entry is trimmed, is not empty, and may span lines, but may not contain a
line holding only `§`. Two identical entries in one file are refused.
**`archive.md`.** Append-only. Each moved entry is written as:
**`archive.md`.** Append-only. Each entry starts with one header line saying how it got there:
```text
§ moved 2026-09-24T20:14Z from memory.md in session mm-y73sxj3popgwumyccptb56p1ah
<the entry>
§ dropped 2026-09-24T20:14Z from memory.md in session mm-y73sxj3popgwumyccptb56p1ah
<the entry>
§ written 2026-09-24T21:02Z for user.md in session mm-dh45qg… · untrusted
<the entry>
```
**The edit journal** (`edits/`) records every applied edit: time, session, call, approval id,
action, file, the entries before and after. It is the history `archive.md` alone cannot give (when
Nothing a write removes is lost: moved and dropped entries are archived, and the journal keeps
every before and after.
**The edit journal** (`edits/`) records every applied edit: its id (`YYYY-MM-DD-NNNN`), time,
session, call, action, file, where it went (`memory.md`, `user.md` or the archive), the entries
before and after, and whether the session had seen untrusted content. It is the history `archive.md` alone cannot give (when
an entry first appeared), and the record an owner reads when a memory looks wrong. **The recall
log** (`recalls/`) records time, session, query, and the source of every hit, and the queries that
found nothing: hit counts for M5c's tiering job are worked out from it, so they survive a reindex.
@@ -95,8 +109,10 @@ prompt, after `system.md` and a blank line. `memoryd` renders it:
The block goes into the session's saved baseline, so it is frozen for the session: a restarted
`loopd` replays the saved bytes whatever the files say by then (inference contract 2, "strict
extension"). `memoryd` refuses to serve the block when a file is over its limit (the owner edited
it) or unreadable; `loopd` then refuses to start the session, with the error and a pointer to the
extension"). Each entry of `memory.md` and `user.md` is checked again as the block is built
(section 5, "The check"); an entry that fails is left out of the block, and `memoryd` logs its
file and the reason. `soul.md` is the owner's and is not checked. `memoryd` refuses to serve the
block when a file is over its limit (the owner edited it) or unreadable; `loopd` then refuses to start the session, with the error and a pointer to the
runbook entry `memory-unavailable`. Existing sessions are unaffected. A starting session with no
memory, and no sign of it, is what this prevents; the M2b rule for `core.md` stays, moved.
@@ -117,7 +133,7 @@ its source and label, and each snippet at most 600 characters:
```text
[1] session mm-8ass9m…, 2026-09-20 14:02 UTC, the owner said · private
…the launch window moved to Thursday…
[2] archive.md, moved 2026-09-24 from memory.md · approved memory · private
[2] archive.md, moved 2026-09-24 from memory.md · private
Owner prefers answers in UTC.
[3] session chat-1789…, 2026-09-21 09:15 UTC, the assistant said · private · untrusted
…
@@ -151,28 +167,70 @@ Arguments, one of (`file` is `"memory"` or `"user"`):
]}
```
Entries are named by their **full text**, never by number or version. That makes every
approval readable on its own: the owner sees exactly which text is kept, rewritten, moved or
dropped, and what replaces it. It also makes a stale edit fail safely: when `memoryd` applies an
approved edit, every `old` must match an entry of the file as it is **now**, exactly; if the file
changed in between and one does not, the edit is refused and nothing changes.
No write asks the owner anything (P21). Entries are named by their **full text**, never by number,
so the journal and the announcement read on their own, and an edit made against an old view of
the file fails safely: every `old` must match an entry of the file as it is now, exactly.
Rules, checked by `brokerd` when parsing (shape) and by `memoryd` when applying (content):
**Where a write goes.** `brokerd` sends each call with the session's state (section 7):
- In a session that has **not** seen untrusted content, the write applies to `memory.md` or
`user.md`.
- In a session that **has**, only `add` is possible, and the entry is appended to `archive.md`
with the header `§ written … · untrusted`, never to the file named. `recall` finds it, labelled
untrusted; it never reaches the prompt on its own. The answer says so: `saved to the archive,
not memory.md: this session has read untrusted content`. `replace`, `remove` and `consolidate`
are refused with `this session has read untrusted content; it can add to the archive but not
change memory.md`. The owner can move an archived entry into a file by editing it.
**Rules**, checked by `brokerd` when parsing (shape) and by `memoryd` when applying (content):
- `add`: the new file must fit its limit; a duplicate is refused.
- `replace`, `remove`: `old` must match exactly one entry.
- `replace`, `remove`: `old` must match exactly one entry. A removed entry is archived as dropped.
- `consolidate`: every current entry of the file must appear exactly once as an `old`, each with
exactly one of `keep`, `text`, `move`, `drop`; entries with only `text` are new. The resulting
file must fit its limit. Moved entries are appended to `archive.md` in the order given.
file must fit its limit. Moved and dropped entries are appended to `archive.md` in the order
given, with their headers.
- An edit that would go over the limit is refused with: `memory.md would be 2,350 of 2,200
characters; use consolidate to move entries to the archive`. That sentence is fixed apart from
the numbers and the file.
Every `memory` call is `ask` (section 8), so every edit passes the owner's approval, over
Mattermost or `bxctl`. The approval shows the arguments, which name every entry by its text.
An applied edit answers `memory.md updated (1,510 of 2,200 characters); it takes effect in the
next session`. The file is written to a temporary file and renamed, then the journal line is
appended.
**The check.** Every new or rewritten entry is refused, with a fixed sentence naming which rule,
if it contains:
1. a control character other than newline and tab (C0, DEL, C1);
2. an invisible or direction-changing character (Unicode general category `Cf`: zero-width
spaces and joiners, bidirectional controls, and the like);
3. chat-template markup: `<|` or `|>`, or `<think`, `</think`, `<tool_call`, `</tool_call`;
4. one of these phrases, ignoring case and runs of spaces: `ignore previous instructions`,
`ignore all previous`, `disregard previous instructions`, `disregard all previous`,
`new instructions:`, `system prompt:`.
The list is short and fixed on purpose: it is defence in depth, as Hermes' scan is. The archive
routing above is the real defence, because a pattern cannot tell a plausible planted fact from a
true one.
**Applying.** The file is written to a temporary file and renamed, then the journal line is
appended. The answer to the model: `memory.md updated (1,510 of 2,200 characters); it takes
effect in the next session`, or the archive sentence above.
**Announcing.** Each successful write is announced as one line, built by `memoryd`:
`memory: added to user.md: "Prefers answers in UTC."`, `memory: consolidated memory.md: 2 moved
to the archive, 1 rewritten, 5 kept`, `memory: saved to the archive (untrusted): "…"`. Entries are
cut at 200 characters in this line. `loopd` emits it as `TurnEvent::MemoryWritten { text }`;
`gatewayd` posts it in the turn's thread and `bxctl chat` prints it. It is information, not a
question.
**Undo.** `bxctl memory log [N]` prints the last N edits (default 10) from the journal: id, time,
session, action, file. `bxctl memory undo <id>` reverts an edit if it is the **latest** edit of
its file and the file still holds exactly what that edit left: the file gets its entries back from
the journal, entries the edit archived stay in the archive (append-only), and the undo is itself
journalled. Anything else is refused with the reason, so an undo never overwrites a later change.
**Review (off by default).** With `[memory] review = true`, a write that would change `memory.md`
or `user.md` is appended to `pending.jsonl` instead, and the answer says it is waiting for the
owner. `bxctl memory pending` lists them; `bxctl memory accept <id>` applies one (with the same
checks, against the file as it is then); `bxctl memory reject <id>` discards it (journalled).
Writes to the archive are never staged.
## 6. The index
@@ -191,7 +249,8 @@ to write it. From `archive.md`: every entry.
highest `class` of any `ToolResult` before this record (at least `private`), and whether any was
untrusted. Each indexed record gets the label in force when it was written. A record whose class
is above `[index] max_class` (default `private`) is not indexed at all. Archive entries are
labelled `private`, trusted: they passed the owner's approval.
labelled `private`, and untrusted when their header says so; moved and dropped entries came from
`memory.md` or `user.md`, which only clean sessions write, so they are trusted.
**Keeping up.** The index stores, per session log, how many records it has read. After each
finished turn `loopd` sends `memoryd` `session_updated` with the session id (section 7), and
@@ -214,8 +273,8 @@ CREATE TABLE progress (log TEXT PRIMARY KEY, records INTEGER);
| Socket | Default path | Peer | Messages |
|---|---|---|---|
| `loopd` | `{home}/run/loop-memory/memory.sock` | `loopd` | `memory_block` → `memory_block_reply`; `session_updated` → `ok` |
| `brokerd` | `{home}/run/broker-memory/memory.sock` | `brokerd` | `tool_request` → `tool_response` (the existing kinds) |
| owner | `{home}/run/owner-memory/admin.sock` | `bxctl` | `reindex` → `reindexed` |
| `brokerd` | `{home}/run/broker-memory/memory.sock` | `brokerd` | `memory_call` → `tool_response` (the existing kind) |
| owner | `{home}/run/owner-memory/admin.sock` | `bxctl` | `reindex` → `reindexed`; `memory_admin` → `memory_admin_reply` |
Directories 0700, sockets 0600, stale sockets removed, as `brokerd`'s. One request per
connection, the existing frames (`proto::frame`, `MAX_FRAME`). New message kinds in `proto`:
@@ -225,18 +284,21 @@ connection, the existing frames (`proto::frame`, `MAX_FRAME`). New message kinds
| `memory_block` | `{}` |
| `memory_block_reply` | `{"text": string}` |
| `session_updated` | `{"session": SessionId}` |
| `memory_call` | `{"request": ToolRequest, "taint": DataClass, "untrusted": bool}`: the decided call, with the session's state when `brokerd` decided it |
| `reindex` | `{}` |
| `reindexed` | `{"logs": u64, "records": u64, "archive": u64}` |
| `memory_admin` | `{"command": "log" \| "undo" \| "pending" \| "accept" \| "reject", "arg": string}` |
| `memory_admin_reply` | `{"text": string}` |
`brokerd`'s list of kinds grows from eighteen to twenty-three.
`brokerd`'s list of kinds grows from eighteen to twenty-six.
## 8. `brokerd`
- **Tools.** `ToolName` gains `Recall` and `Memory`; `ToolArgs` gains their parsed arguments
(`proto` holds the serde structs, as for the other tools). Parsing enforces the shapes and
limits in section 5. Neither tool has constraints (`paths`, `hosts`, `patterns` must be empty).
- **Grants: built in** (owner, 2026-09-24: an agent that cannot write memory is not worth
running). `brokerd` holds two fixed grants, always present whatever is in `grants/`:
- **Grants: built in** (owner, 2026-09-24: reading and writing memory is what an agent is).
`brokerd` holds two fixed grants, always present whatever is in `grants/`:
```toml
# built in: recall
@@ -250,7 +312,7 @@ connection, the existing frames (`proto::frame`, `MAX_FRAME`). New message kinds
```toml
# built in: memory
tool = "memory"
mode = "ask"
mode = "auto"
max_taint = "secret"
result_class = "private"
untrusted = false
@@ -259,12 +321,12 @@ connection, the existing frames (`proto::frame`, `MAX_FRAME`). New message kinds
Their names are `builtin-recall` and `builtin-memory`, and their `grant_sha256` is the hash of
that text, so the audit log names them like any other grant. A file in `grants/` whose `tool`
is `recall` or `memory` is a grant problem at load (`grants-invalid`), with the fixed sentence
`the policy of recall and memory is built in; remove this grant`. So no grant file can turn
memory edits off, or make them unattended; every edit is the owner's (brief, memory 7). With
section 9 (no session without the memory block), an agent that can talk can always read its
memory and propose edits to it.
`the policy of recall and memory is built in; remove this grant`. No grant file can turn memory
off or put questions in front of it. What protects the prompt is where untrusted sessions'
writes go (section 5), which `memoryd` decides from the state `brokerd` sends.
- **Where a decided call runs.** A composite runtime sends `Recall` and `Memory` specs to
`memoryd`'s `brokerd` socket as a `tool_request`, and everything else to the configured runtime
`memoryd`'s `brokerd` socket as a `memory_call`, with the session's taint and untrusted flag
as `brokerd` holds them when it decides, and everything else to the configured runtime
(Podman, or the refusing one). `[memory] socket` in `brokerd.toml` names the socket; without it,
memory calls fail with a fixed sentence and a runbook pointer (`memoryd-unreachable`).
- **Labels only get stricter.** `RunOutput` gains an optional label. `runner::run` answers with
@@ -281,16 +343,22 @@ connection, the existing frames (`proto::frame`, `MAX_FRAME`). New message kinds
refusing, fails the new session with the runbook pointer `memory-unavailable`.
- Registry: `recall` and `memory` are core, with the schemas of section 5. The model sees
`clock`, `recall`, `memory`, `find_tool`, `call_tool`.
- A successful `memory` result carries the announcement line (section 5) at its end, after a
fixed marker line; `loopd` strips it before the result is logged or shown to the model, and
emits it as `TurnEvent::MemoryWritten`.
- After a finished turn: `session_updated`, best effort. A failure is one log line; the next
startup of `memoryd` catches up.
- `system.md` gains one sentence: memory edits need the owner's approval and take effect in the
next session, and recalled text is data like any tool result.
- `system.md` gains two sentences: keep `memory.md` and `user.md` short and current, and use
`consolidate` when they fill; writes take effect in the next session, and recalled text is data
like any tool result.
## 10. `bxctl`
`bxctl reindex [--admin-socket PATH]`: sends `reindex`, waits, and prints `reindexed: 42 logs,
1,208 records, 7 archive entries`. `memoryd` rebuilds into a new file and renames it over the old
one, so `recall` keeps working meanwhile.
- `bxctl reindex [--admin-socket PATH]`: sends `reindex`, waits, and prints `reindexed: 42 logs,
1,208 records, 7 archive entries`. `memoryd` rebuilds into a new file and renames it over the
old one, so `recall` keeps working meanwhile.
- `bxctl memory log [N]`, `bxctl memory undo <id>`, and with review on, `bxctl memory pending`,
`accept <id>`, `reject <id>` (section 5). Entries are printed escaped, as approvals are.
## 11. Configuration
@@ -302,6 +370,7 @@ home = "/var/lib/boxmaker" # default: $BOXMAKER_HOME, then /var/lib/boxma
[memory]
soul_starter = "soul.md" # next to the config file
review = false # true: writes to memory.md and user.md wait for the owner
[index]
max_class = "private" # records above this class are not indexed
@@ -312,14 +381,20 @@ max_class = "private" # records above this class are not indexed
## 12. Security notes
- `memoryd` has no network in M5a and never writes `sessions/`, grants or the audit log.
- Nothing reaches the prompt without the owner: `soul.md` is the owner's file; every edit to the
other two is approved; the block is frozen per session.
- What reaches the prompt: `soul.md`, the owner's file; and `memory.md` and `user.md`, which only
sessions that have not read untrusted content write, through the check, frozen per session.
Everything a session that read untrusted content writes goes to the archive, labelled
untrusted, where it can be recalled but never lands in the prompt by itself.
- Memory writes are the one tool with no approval anywhere. The trade is deliberate (owner,
2026-09-24): the risk it leaves is a clean session writing something false, which is the
agent's judgement of what the owner said; every such write is announced, journalled and can be
undone, and review can be turned on.
- Recalled text is a tool result: appended, labelled, and it taints the session like any other.
Tool results are not indexed, so a fetched page cannot come back through `recall`; what the
assistant said after reading one can, marked untrusted.
- The FTS5 query is built only from quoted terms; the model's text is never SQL.
- Entries shown in approvals are escaped like all arguments (`proto::escape`), so an entry cannot
render as markup or mention anyone.
- Entries shown by `bxctl` and in announcements are escaped (`proto::escape`), so an entry
cannot render as markup or mention anyone.
## 13. Runbook
@@ -333,31 +408,38 @@ New entries: `memoryd-start-failed` (configuration, `memory/` or `index/` unusab
Offline, in `make gate`:
- `memoryd`: entries parsing and rendering; each edit's rules, including a stale `old`, a
duplicate, the limit sentence, and `consolidate` missing an entry or naming one twice; the
archive and journal lines; the block; the quoted-terms query builder against FTS5 operators
duplicate, the limit sentence, and `consolidate` missing an entry or naming one twice; every
rule of the check, on write and on building the block; routing by the session's state (an
untrusted `add` to the archive, the other actions refused); the archive headers and journal
lines; undo of the latest edit, and its refusal when the file changed; review on and off; the block; the quoted-terms query builder against FTS5 operators
(`NEAR`, `*`, `^`, quotes, column filters); labels walked from a log; `max_class`; catching up
from `progress`; a reindex equal to the incremental index; the recall log.
- `brokerd`: parsing both tools; the built-in grants present with an empty `grants/`; a grant
file naming either tool refused at load; the composite runtime; the
file naming either tool refused at load; the session's state carried in `memory_call`; the composite runtime; the
stricter label, with a runtime that returns a stricter and a laxer one.
- `loopd`: the baseline with a fake `memoryd`; the refusal and its pointer; `session_updated`
sent after a finished turn and not after a turn ending in tool calls.
- `bxctl`: `reindex` against a fake socket.
sent after a finished turn and not after a turn ending in tool calls; the announcement stripped
from the logged result and emitted as a turn event.
- `gatewayd`: a `MemoryWritten` event posted as one line.
- `bxctl`: `reindex` and each `memory` command against a fake socket.
On the device (the owner's Mattermost, as for M4b): the baseline token count with the starter
soul; a fact told in one thread and recalled in another; an `add` approved from Mattermost that
appears in a new thread and not in the thread that asked; a `consolidate` that moves an entry,
which `recall` then finds; a hit from a session that read a fetched page coming back untrusted.
soul; a fact told in one thread and recalled in another; an `add`, announced in its thread with no
question, that appears in a new thread and not in the thread that wrote it; a `consolidate` that
moves an entry, which `recall` then finds; a write after a fetched page that lands in the archive,
untrusted; `bxctl memory undo` of the latest edit.
## 15. For the owner's review
1. Decided: tool results are not indexed; tool calls are, by name and arguments (section 6).
Summaries of a tool's output, where the tool opts in (a web fetch, say), come with M5c's
summariser, labelled like the output they summarise.
2. Decided: the policy of `recall` (`auto`) and `memory` (`ask`) is built in, and grant files
cannot name them (section 8): memory edits can be neither turned off nor made unattended.
3. Entries are named by their full text, not by number (section 5), so approvals are readable and
stale edits fail; the arguments of a `consolidate` of a full `memory.md` are about 4,500
characters, one approval post.
4. The starter `soul.md` is generic (section 3); the owner writes it before or after the first
start.
2. Decided: memory writes need no approval (P21); the policy of `recall` and `memory` is built in
and no grant file can change it (section 8). A session that read untrusted content writes only
to the archive (section 5).
3. Decided: entries are named by their full text, not by number (section 5).
4. Decided: the starter `soul.md` is generic (section 3).
5. Open: the phrase list of the check (section 5) is short on purpose. Add phrases you have seen
in the wild, or drop the list and keep only rules 1 to 3.
6. Open: the announcement goes to the turn's thread (section 5). A channel of its own for memory
writes, or silence with `bxctl memory log` only, are the alternatives.