Memory now has one switch, On or Off, in the Memory tab of the account
settings. Off means nothing from the notes is put in the prompt, the
model has no memory skill, tool or note kind, nothing is proposed or
saved (also by triggers and other runs that start by themselves), and
the model is told in one line that memory is off. The notes stay and can
still be read, exported and deleted.
A chat can be private: the Memory row of its Tools menu is the memory
switch group, stored with the chat. The narrowest level wins: chat,
agent, person, server. A saved agent cannot widen the choice of the
person. The tools list says which level is off (memory.code, blocked_by,
fix); the chat shows a chip and one line, and the agent form shows the
person's switch as a read-only line.
New API: GET and PUT /api/chats/memory for the switch and the save level,
and GET /api/chats/memory/export for a zip of the notes. The Memory tab
gains the switch, the three save levels, Export and Forget everything.
Assisted-by: Claude:claude-sonnet-5-5
The manifest said its text in English only. An add-on can now translate
what the host shows to people, and the library list shows it before the
add-on is installed.
The title and the description use the language maps of ui.title and
ui.description, which every host already reads. A new optional field,
i18n, translates the connector kinds by language tag. It stays under api
2 and its set of fields is closed: the title and the description of a
declared kind, and only where the English kind has them. Tool
descriptions, schemas and skills are for the model and stay English.
Tags follow the language map rule (at most 16, no English), each string
has the limit of its English field, all translated text has at most
12 000 characters, and markup, control characters and direction
overrides are refused. A host that does not know the field refuses the
manifest, as it refuses any unknown field. The field is part of the
manifest, so the package digest and the signature cover it.
The catalog generator copies the translated title and description of
the manifest into each library entry (ui.titles, ui.descriptions), the
host validates them like the manifest values, and a library search
matches the English and the translated text. The apps list and the
connector kinds return the text of every language and, when the request
names a language (lang, else Accept-Language), the resolved text beside
the English fields: language, localizedTitle and localizedDescription.
No existing field changes.
Every add-on on master now has its title and description in Italian,
German, Spanish, French, Brazilian Portuguese, Simplified Chinese,
Korean and Indonesian, and the chat connectors carry their kind text in
the same languages. The packaging tests require that, and that a
translation has the fields of the English text it translates. The names
of the add-ons that provide a role are matched in their English title
only: a translation is often an ordinary word of its language.
Assisted-by: Claude:claude-sonnet-5-5 go
Search read the newest 500 chats and scanned them in memory, so older
chats dropped out silently. The chat store now keeps a search index of
the text of user and assistant messages: FTS5 with the trigram
tokenizer on SQLite, a pg_trgm GIN index on PostgreSQL, and a scan of
the same table when neither is there. The file store scans its
in-memory transcripts for up to two seconds, newest chat first, and
says when it stopped early.
The index row is written in the transaction that writes the entry and
removed in the one that deletes the chat or truncates the transcript.
A query joins the chat row for the owner, so one user never sees the
chats of another, and sub-agent chats stay out. Messages written before
the index existed are added in the background with a resumable cursor
and progress in the log; startup does not wait for it.
GET /api/chats/search keeps its fields and adds the role, time, match
count and highlight offsets of the hit, and filters for agent, origin,
role and date range. A phrase in double quotes is one word. Hits rank
by match quality and recency. group=message lists every matching line.
Assisted-by: Claude:claude-sonnet-5-5
The Memory add-on only declared three index files, two skills and a save
setting; the host did the rest. Without the add-on a memory proposal
worked but nothing was read back, and removing the add-on silently reset
the save level.
The host now loads the three indexes (user, agent, conversation) with
the same caps and data wrapping, serves context_read, ships the two
skills as built-in skills with the same text, and keeps the save level
as its own per-person setting (ask, auto, never; default ask). The code
sits behind a small MemoryProvider interface.
The add-on stays installed and harmless: its index files and skills are
ignored while the host serves memory, so nothing loads twice. A person
who chose a level on its page keeps it until they choose in the host
setting; auto still needs the add-on to run. A person who disabled the
add-on and never chose keeps memory off. Saved agents still read notes
only when their add-ons list names memory, and messaging chats still
ask before saving.
A person with no note and no add-on gets no extra block in the prompt.
Assisted-by: Claude:claude-sonnet-5-5
The texts for a blocked capability put a noun phrase into slots that
need a verb, so some read badly ("was not given writing to other
channels", "before reminders and scheduled tasks"). Each capability now
has an action phrase in the infinitive and a noun phrase. Every code is
a complete sentence built from the action phrase, for the owner and for
the sender and the model, and most say what to change.
The note for the model now sets whole sentences apart with a space and
asks the model to relay the reason plainly. A table-driven test renders
every code for every capability and checks grammar, length and that no
internal name leaks. Codes, JSON shapes and behaviour do not change.
Assisted-by: Claude:claude-sonnet-5-5
The word "Memory" named collections of documents, notes about the
person, and RAM in the same navigation. Collections are now Knowledge:
the sidebar item, the page title and text, the Home action, the agent
form section and the chat tool group. The tool group "Memory and
profile changes" is now "Suggested changes", since it is the generic
ask-first mechanism for notes, rules, skills, pages and the dashboard.
The Settings section "Memory and models" is now "Models in memory".
Only display strings change. Ids, routes, locale keys, group ids,
grants and stored data stay as they are. All nine locales are updated;
the docs name the Knowledge page and say which search the default
engine does.
Assisted-by: Claude:claude-sonnet-5-5
Chats and agents can read a Home Assistant instance and, if the person
allows it, switch things in it.
Read tools: server summary, areas, entity search, state, history,
logbook and the catalogue of callable services. Results are bounded,
marked as untrusted data, and stripped of pictures, access tokens,
secrets and locations.
Write tools: call-service for quiet services (lights, switches, input
booleans, media transport) follows the write mode of the connection;
call-physical-service always needs a confirmation. A deny-by-default
allowlist decides which services a model may call. A hard exclusion
list that no setting can lift covers locks, alarms, sirens, valves,
cameras, garage-type covers, restarts, reloads and shell, system and
add-on management. Calls name entities one by one, up to a cap.
Entities can be excluded by domain, area, pattern or label.
One connection is one instance: the token is the credential and the
address is its one allowed resource, so the host refuses any other
address before the add-on runs. The add-on speaks only to that address,
refuses redirects, verifies TLS, and accepts a self-signed certificate
only through a pinned SHA-256 fingerprint.
The policy lives in the add-on data folder and is edited on the add-on
page by an administrator. For that, the host no longer asks a
credential for a uiOnly tool of a credentialed add-on, since the page
has no chat to name a connection.
Assisted-by: Claude:claude-sonnet-5-5 go node
List the add-on in the package workflow matrix, the package checks, the host
operation policy, the catalog template and index, the case inventories of the
contract and packaging suites, and the architecture guard (two links to the
project community server are not the add-on). Document the connector in the
chats guide.
Assisted-by: Claude:claude-sonnet-5-5 go
Let chats and agents query a Jellyfin media server and, when the person
allows it, remote control its players. Read tools: search, libraries,
browse, item details, latest, continue watching, next up, playlists,
sessions, user names and server info. Write tools: control-playback
(play, pause, resume, stop, seek, volume), mark-played, set-favorite and
add-to-playlist. There is no tool that deletes, scans or edits the
library or manages users.
One connection holds one API key (credential kind token) and one server.
The server address is the single entry of the connection's resource
list, so the host checks it before every call and several servers work
through the connection selector. The add-on talks only to that address:
http and https only, no user information or query, redirects refused,
paths built from fixed segments, and the key sent in the Authorization
header only. A self-signed certificate is accepted only when the
address carries #sha256=<fingerprint>; the connection then accepts that
one certificate and TLS checks are otherwise on.
Results are bounded (50 items a page, cut overviews), marked as
untrusted data, and hold no file path, stream or image address. The
write mode off, propose or auto is the host's per-connection setting.
A write is previewed from the server's own records and rejected as stale
when the player or playlist changed before the person confirmed.
The add-on ships a skill, a connection page with a Test button and
approval cards, the package files, the library entry and the docs, and
joins the package workflow and the case inventory.
Assisted-by: Claude:claude-sonnet-5-5 go node
A note, a profile, a skill, a page or a dashboard change that the agent
proposed in a messaging chat could not be confirmed there: a plain "yes"
went to the model, and the tool result sent the person to Studio, which
the connector prompt says they do not read.
The host now writes the question in the chat, with the kind, the scope
and the lines that will be saved, and answers the reply before any model
turn. yes/ok/confirm saves, no/cancel drops, a four-character code picks
one when several wait, and "undo" takes back a save for 30 minutes. Only
a sender with proposals.confirm may answer; the kinds that change what
the agent may do need a paired sender, and in a shared chat only the
sender who asked. The tool result tells the model what the host did, and
the Studio wording stays for Studio chats.
The permission text of "Confirm changes from the chat" now says what it
covers. The proposal source keeps the asking sender.
Assisted-by: Claude:claude-sonnet-5-5
The access view said only that a capability was refused, in free text.
It now reports, for every capability of every kind of sender, the first
link of the permission chain that fails: a closed code, who holds the
link (server, tools, agent, connection, sender), one plain sentence, one
fix as a kind and a stable target, and the links not yet checked. The
chain order is server setting, Tools switch, agent permission, the
permission of the connection, then sender trust. The code is produced
where Authorize and the capability switches decide, so the English
reason and the decision cannot drift apart. The Tools switch of the
agent and the server triggers setting are now switches in that chain.
POST /api/chats/connectors/{id}/access/preview runs the same chain on a
form that is not saved (id "new" works with a kind), optionally for one
kind of sender. It writes nothing, records nothing and never takes or
returns the token.
The held-back note for the model, the tool errors and the trigger tool
reason take their sentence from the same table of codes, in plain words
with no setting names.
Assisted-by: Claude:claude-sonnet-5-5
The block of earlier thread messages listed the asker's own earlier
messages under their name, inside a block that said it held messages
from other people. The model then said it could not see what the asker
wrote. Lines of the asker are now "The person asking (Name)", lines of
the bot "Assistant (you)", and the heading says the block holds the
earlier messages of the thread, including those two, as untrusted text.
The asker is matched by platform id, then user name, never by display
name. A name that imitates a fixed label is shown as "Someone (...)".
Each thread read now logs one INFO line with counts only: messages
returned, lines kept, bytes, truncated, present or empty. A kind that
serves no thread history logs a distinct line.
Assisted-by: Claude:claude-sonnet-5-5
A connection kept one chat per person for good, with no way to ask for
a new one. A person can now send /new (or !new, for Slack, which keeps
messages that begin with a slash for itself) as the whole message.
The host reads the command after the pairing check and before any
model turn, so the agent never sees it and the logs record only the
ids. It drops the link from the person, or from the shared thread or
group, to its chat. The old chat stays in Studio. The link is in the
shared store, so it works with several frontends and for every
connector add-on.
In the chat mode and in a thread of a group, everyone shares the chat,
so only a paired sender may reset it. The thread mode answers that
every message already starts a new chat. A chat with a running turn is
left alone and the sender is asked to repeat the command later.
Telegram did not treat /new@BotName as addressed to the bot in a
group; it does now, and keeps the text. The connection page gets one
help line under Conversations.
Assisted-by: Claude:claude-sonnet-5-5
The spec "a turn that stops beating is reaped, one that keeps beating
is not" beat the live turn once and then made several store calls
before the second reap with a 50 ms threshold. Under -race on a loaded
runner those calls took longer than 50 ms, so the live turn counted as
stale and the second reap found it.
Use a 300 ms threshold, sleep twice that for the stale turn, and beat
the live turn again right before the second reap. The assertions keep
their meaning and the store is unchanged.
Assisted-by: Claude:claude-sonnet-5-5
The platform e2e spec "ends a tool call that outlasts the 60 second
limit" has failed on master since connector tool calls got the MCP tool
timeout (360 s by default) instead of the connector package default of
60 s. The fixture tool sleeps 90 s, the call is no longer cut at 60 s,
and it returns 200 with "slept". The new limit is intended, and a
standalone server has no flag to shorten it, so the spec can only pass
by waiting more than six minutes.
Move the check to the studio service tests: a connector service with a
one second call time limit, and an app tool that sleeps past it, must
end with ErrMCPAppFailed and no credential in the text. Remove the e2e
spec.
Assisted-by: Claude:claude-sonnet-5-5
A chat with no add-on that runs commands could not tell a person
whether the server's built-in bash was off or simply not selected.
The add-on list now carries commands_host_exec (off, ask or on), a
read-only copy of the chats_host_exec setting as it applies now.
Assisted-by: Claude:claude-sonnet-5-5
The gate behind the auto start of a trigger with acts asks about a
capability that is not sensitive, with the Sensitive mark. Pin that an
allowlisted sender alone is refused and a paired one passes.
Assisted-by: Claude:claude-sonnet-5-5
A trigger proposed from Slack or Telegram was always read only, and
the person had no way to let its runs act except to edit the trigger
on the Triggers page afterwards.
The approval of a binding now has an acts list, the act.* permissions
that a proposal from its chats may carry. It is empty by default, so
nothing changes until the owner sets it. It is stored in the same
approval JSON (no migration), checked against the closed list in the
API, and always limited to what the agent holds.
The acts of a proposal are what the model asks for, within the list of
the binding and the grants of the agent. The rest is dropped and
named in the proposal text, and the model is told which acts it may
ask for and to ask the person first when the job changes things. A
read only proposal now says what the person can do next.
For a proposal with acts, auto_readonly still asks. auto starts it at
once only when every act is allowed for the binding, nothing was
dropped and every sender of the turn is paired; otherwise it asks. A
trigger made without asking also keeps to the acts of its binding at
run time.
Assisted-by: Claude:claude-sonnet-5-5
A binding with its capabilities ticked still gave nothing to the
owner's own allowlisted account: Authorize wanted a pairing or a role
even for capabilities that are not sensitive, so the agent had no
reminder, memory or talk-back tool.
A sender whose ID or @name is an entry of the allowlist now holds every
capability that is not sensitive, within the ceiling of the binding and
while no switch is off. The decision says Via "allowlist". A sensitive
capability still needs a pairing, and a role never grants it.
The entry of a channel or group lets its members talk to the bot but
does not trust them: the trust is matched against the sender only, and
a binding that accepts anyone lists nobody. Such a sender still gets
capabilities from a pairing or a role. The capability gate takes the
handle of the sender for the same match, and a sensitive request
ignores the allowlist as it ignores roles.
When a tool is held back for a missing right, the turn now carries a
short note with the reason in plain words and tells the model to relay
it, instead of saying that it has no such tool. The reasons are kept
in the grant of the turn, so a worker reports the same words. The
access view lists a "listed" sender apart from one let in through a
channel.
Assisted-by: Claude:claude-sonnet-5-5
A model could write a paperclip line with a made-up file name and
never call attach_file. Nothing was sent and the sender got no note.
When the answer of a connector turn has a paperclip followed by the
name of a sendable file, and no file went and no other note was
sent, the host now adds one plain note. The answer is not changed.
Prose that names a file and a paperclip alone do not match.
The attach_file description and the connector guidance now say to
use the exact path the tools showed, and never to write a file name
or a paperclip line by hand. A spec covers a folder name with a
space.
Assisted-by: Claude:claude-sonnet-5-5
AuthorizeConnector had no case for the op history, so the default
branch refused it. A mention in a Slack thread then logged "connectors:
forbidden" and the first turn of the new chat got no thread block.
History now follows the rules of a send: the binding is the whole
authority, the token comes from owner storage and a disabled binding
keeps it. Only the poll stops with the binding.
The studio specs ran over a transport that skipped the scope check, so
none of them saw the refusal. The rig now authorizes every call with the
real manager, and a new spec table pins each op of the contract.
Assisted-by: Claude:claude-sonnet-5-5
The turn helper of the talk-back specs ended its wait on the first text in
the chat. That text is the working message that goes out when the turn
starts, so a spec could read the tool results before the tool had run.
The refused-channel spec failed that way under load and the race
detector. Count only the replies that are not the working message.
Assisted-by: Claude:claude-sonnet-5-5 go
The text of the report reaches the channel before the access row is
written, so reading the rows once right after the text was a race.
Assisted-by: Claude:claude-sonnet-5-5 go
The coverage run and the no-skips run share one database. The migration
claim left its rows behind, so the second run found the claim taken and
two specs failed. Wipe the two tables the setup left out.
Assisted-by: Claude:claude-sonnet-5-5 go
A memory note is still a proposal that the person confirms by default.
An add-on can declare the user setting with the role memory.save-mode,
and the person then picks: ask first, save it and tell me, or never.
With auto, a note from the person's own Studio chat is applied at once
through the whole proposal path (grant, gate, checks, rate limit, audit)
and confirmed as the person, so it is one change set that revert_change
and the Undo on the chat restore. The chat shows "Saved to memory" with
Undo and no confirmation card. The record carries auto: true and the
audit says the write was automatic.
Safety rules live in the host. A connector chat always asks, a trigger or
job has no tool, a saved agent cannot widen the choice, and a failure to
read the setting asks. At most 5 notes are saved unasked in a turn; a note
over 4 KiB or one that looks like it holds a secret becomes a normal
proposal. With never the memory kind is hidden from propose_change and a
stray call is refused. Only memory reads the setting.
Assisted-by: Claude:claude-sonnet-5-5 go node playwright
A chat now enables a set of connections for an add-on whose tools run with
a credential, and the model names one of them in each call. This applies to
every add-on that declares a credential in its manifest.
- The Tools menu row of the add-on lists the connections with a checkbox
each. The saved agent names a set too (an empty set means each chat
chooses). A chat that a connector binding started keeps that binding
alone. With one connection, it is the set. With none enabled, the tools
are not offered.
- With one connection enabled the tools are unchanged. With two or more,
the host adds a required `connection` argument to every tool of the
add-on in the definitions it gives the model: an enum of the connection
names (the display name folded to lower case, digits and hyphens; the
description keeps the display name). The add-on never sees the argument.
- The name is checked against the set the chat has now, not against all the
connections of the person. A missing or unknown name returns a sentence
that lists the allowed names, and nothing is sent. A write tool never runs
without the name. Text in a tool result cannot widen the set, and the
owner checks the chosen binding against the set again.
- The resource list of the named connection is the only one that applies.
Results, the approval card and the text for a pending change name the
connection; the audit line has the binding ID.
- A connection name must be unique per person and add-on among the
credential-only connections, compared by name fold. Older duplicates get
a numeric suffix in the model's list.
- The stored chat, draft and agent values move from one binding ID per
add-on to a list of IDs. A single string is read as a set of one in the
stores and in every request, so older data and clients keep working.
PUT /api/chats/{id}/connections takes `connections`; the older
`connection` field still works.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt ginkgo node eslint playwright swag
A chat never uses a connection that the host picked for it. For an add-on
whose tools run with the token of a connection (such as GitHub):
- With one connection, it is selected. With several, the Tools menu row
of the add-on has a Connection selector, and until the person picks one
the tools are not offered and the row says so. With none, the row says
so and links to the Connectors page. The choice is stored with the chat
and its draft, like the other switches, and follows a fork, a
sub-agent and the runs of a trigger the chat proposed.
- A saved agent can name its connection per add-on (agent form field, API
field connections). Its chats use it, show it read only and cannot
change it. An agent with no choice follows the same rule as a chat. A
chat that a connector binding started keeps using that binding.
- CredentialFor takes the selected binding and uses only it, so the
resource list of that binding is the only one that applies. A call with
no selected binding is refused with the code select_connection and a
note the model can relay. The first-match code (connectors.Pick) is
removed.
- Tool results and proposal cards say which connection acted, for example
"as Work, repository acme/widgets". A proposal is applied with the
connection it was previewed with. The credential use is logged with the
binding ID.
Contract: the credential slot takes an optional resourceLabel (the word
for the resource in that line). The agent export file does not carry the
connections, since a binding ID belongs to one server.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt ginkgo node playwright protoc swag
A connector kind can declare agentOptional. A binding of such a kind may
have no agent: it holds the token, the resource list and the write mode
for the tools of a chat, and never receives or answers a message. The
GitHub kind declares it, so a person no longer makes a throwaway agent
only to store a token.
Without an agent the host refuses every setting that needs one:
listening, accepting senders, an allowlist, a notification chat,
talk-back channels, permissions, pairing and trigger approval. The
places that read the agent of a binding handle the empty value:
inbound routing, the polling loop, notifications, pairing, the capability
check (deny by default) and the grant switch. A kind that does not
declare the key still needs an agent on every binding.
The binding page makes the agent field optional for such a kind with
help text, and the list shows "Credentials only".
Assisted-by: Claude:claude-sonnet-5-5 go gofmt ginkgo node protoc swag
A poll that began before the add-on stopped could still be counted, so
the spec for a binding without its add-on failed at times. Take the
count after a short quiet and require that it does not grow.
Assisted-by: Claude:claude-sonnet-5-5 go ginkgo
Bindings saved before permission enforcement have an empty capability
list, so file upload, triggers, memory and talk-back stay denied until
the owner sets them. Give those bindings the set that "Allow the usual"
grants, one time.
An empty list is stored the same whether the owner never chose or
cleared every switch, so the store keeps a migration record instead of
a flag on each binding. The record is claimed atomically with a lease,
so several frontends starting together run it once, and a finished
record is never claimed again: later empty lists are the owner's choice.
Each binding is written only while it still has no capabilities, so a
save made at the same moment is not overwritten.
Only bindings whose kind is running and lists permissions are changed;
sensitive capabilities are never granted. One log line per binding and
one summary. LOCALAI_SKIP_BINDING_CAPS_MIGRATION skips it.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt protoc
A trigger run with nobody present only reads unless its trigger holds
act.* grants, and until now only the REST API could set them. An agent
that proposes a trigger can now ask for acts too:
- propose_trigger takes an optional acts list. The names must be
act.* grants the agent of the chat holds; anything else is refused
and nothing is stored. A read_only proposal cannot carry acts.
- The proposal stores the list. Confirming it, in Studio, by a plain
yes or by confirm CODE, grants exactly that list. No tool changes
the acts of a trigger afterwards. Two proposals that differ only in
their acts are kept apart.
- The text of a proposal in a messaging app now has a Permissions
line, in plain words with sensitive classes marked: read only, or
what a run may do. The confirmation repeats it. A proposal that asks
for acts is never set up without a yes, whatever the approval mode.
- The trigger REST object gains acts_held, read_only and
read_only_why, computed with the same rules as the run, so a client
can show why a run only reads.
Assisted-by: Claude:claude-sonnet-5-5 go ginkgo
A new chat with the sandbox switched off in the composer showed no
"switched off" notice. GET /api/chats/draft/addons took only the model
and the agent, so it always reported the default commands state.
Accept the draft's switches as comma-separated query parameters
(addons_off, builtins_off, builtins_on), check them as the create request
does, and apply them to the draft record before the commands state is
worked out. The composer sends its choices each time an add-on or a
built-in group is switched, so the notice follows the switch.
Assisted-by: Claude:claude-sonnet-5-5 go-test playwright
An agent with no grant list keeps its non-sensitive connector features
(trigger tools, talk-back, files from senders). Granting any one built-in
makes the list the whole answer, so the others stop working. The editor
had no grants list at all, so nobody could see this.
Add a Permissions section to the agent editor with the declared grants.
When the first grant is ticked on a saved agent that had none, a notice
names the connector grants that would turn off and a button adds them.
Also fix the files.upload text. The gate enforces it on files a sender
attaches to a message, not on the agent sending files out, so the grant
is now "Receive files".
Assisted-by: Claude:claude-sonnet-5-5 go-test playwright
Connecting a connector must reach both MCP transports. Adds specs that resolve a
Streamable HTTP server promptly and a legacy SSE server, through the same
mcptransport.Connect path that agent runs use.
Signed-off-by: rossg <ross@golder.org>
(cherry picked from commit 4feaf9db39)
(cherry picked from commit fab3cfa21c13747e03115d7cc39f2466962f16e8)
A trigger run or an agent job has nobody to approve a step, and so far
the approval mode auto_readonly and the act.* grants were only recorded.
The prompt was the only thing that kept such a run from acting.
Every tool call of an unattended run now goes through one check at the
call. A tool that only reads runs. Any other tool needs the grant of its
class, and the check fails closed:
- act.files for write_file, edit_file, move_file, delete_file and the
built-in bash;
- act.addons, or act.addons:<id>, for add-on tools whose declared kind is
not read, plus act.spend when the add-on gates the tool with a
permission;
- act.spend for MCP tools.
A trigger holds its new acts list within the grants of its agent. A
trigger that declared read only runs, or that a binding in auto_readonly
mode made without asking, is refused every non-read tool whatever it
holds. A binding that cannot be read counts as auto_readonly. A job
holds the act grants of its agent. Sub-agents follow their parent, and a
saved agent cannot be started from an unattended run.
A refused call answers with a plain sentence the model can pass on, not
an error, and writes one audit log line. Triggers set acts through the
REST API; a model never can. A trigger whose plan names a file it writes
keeps act.files.
Assisted-by: Claude:claude-sonnet-5-5 go ginkgo swag
The Skills page asks for the skills of the saved agents with
agents=true and shows them with the agent's name. A skill of the
agent layer says who proposed it, can be removed (to the Trash) and
is not offered to other agents. The list of a plain chat or of the
agent form never carries them.
Assisted-by: Claude:claude-sonnet-5-5 go node
A saved agent keeps its own skills in LocalAI/agents/<agent>/skills.
Only a chat that runs that agent sees them. They sit between the folder
and personal layers, so an agent's version of a skill wins for that
agent, and they count as granted without being listed in the agent's
skills.
An agent writes one by a skill change proposal, as before; the new
layer only reads them. A proposal for an agent skill cannot take the
name of a core skill or of a document slot (read from slots.json).
When writing-skills loads in a chat of a saved agent it starts with
"Do not write files. Call propose_change with kind skill.", and the
skills.write grant makes it visible to the agent.
The Skills list gains the agent layer, with the agent's ID and name,
and a skill can be read or removed (to the Trash) with the agent
query parameter.
Assisted-by: Claude:claude-sonnet-5-5 go swag
The draft tools and add-on lists, and the tools list of a stored chat,
read the execution status without an execution scope. The host logged
an error and reported the status as not selected. They now run in a
scope; a draft has no conversation, so its status says "draft".
In the Home composer, switching a built-in tool group on or off reset
the commands state to the loading value and never read it again. The
composer then showed the unknown warning for good. The Hide button also
did nothing for that state, because it needed a binding identity that
the loading value does not have. A draft now keeps the state the list
gave it, and Hide works for any notice that is shown.
Assisted-by: Claude:claude-sonnet-5-5 go ginkgo node
A connector add-on no longer carries file bytes in the connector
envelope. The ops media and send_file need a transfer handle, and the
gateway refuses a call to them without one, or for a kind that declares
no files.
- The manifest field files.transfer, the inline file of a request and
of an answer (ConnectorFile) and the 1 MiB cap
(MaxConnectorFileBytes) are removed. The 2 MiB limit of a connector
call between hosts drops to 1 MiB, because it holds text and no file.
- A kind that declares no files moves no file: the host answers a sender
who attaches one with a plain sentence and does not offer attach_file.
A direction left out of a declared files block still gets the host
ceiling.
- The inline layer is gone from the file limit, from the API
(inline_bytes) and from the Files section of the connector page.
- The Telegram and Slack add-ons keep only the streaming paths. The
tests moved to the handle path, and the studio test rig gives every
connector add-on a table of handles.
Assisted-by: Claude:claude-sonnet-5-5 go test node
No release carried add-ons, so nothing reads older add-on data or
manifests. Delete the code that did.
- Connectors: the kind and the add-on of the Telegram connectors that
core once carried are gone (LegacyAddon, KindTelegram, the stored
binding normalization, the platform titles). A binding needs a kind,
and its add-on is the one that serves the kind. The conversion of the
old agent pool lists every connector as dropped instead of making a
Telegram binding, and its notes say so.
- The library index reads only the id field of an entry, not addonId.
- A host control that has a capability block (the network policy) is
attached by that block alone, not by a tool that replaces bash.
- The Office create tools no longer answer for the argument names of an
earlier schema, and the identity file note about old addresses is
gone.
- The known debt of the provider-name guard drops the entries of the
conversion and of the resource listing shim, and its ceiling goes
from 13 to 5.
Assisted-by: Claude:claude-sonnet-5-5 go test
For a kind that declares files.transfer, the connector manager gives the
add-on a handle and not the bytes, in both directions.
Inbound, the add-on streams the file to the host. The chat upload then
reads it from the blob store, so a 6 MB PDF reaches the chat. Outbound,
the manager checks the access of the chat to the file, gives the
process that runs the add-on only the hash and the size of the content,
and the add-on pulls the chunks. The file limit of such a kind has no
inline cap: it is the smallest of the host ceiling, the ability the kind
declares and the limit the owner set for the bot, and the sentence names
the number and the layer. A kind without transfer keeps the 1 MiB inline
cap, and its sentence says so.
Checks: the files of one message may add up to three times the limit;
the media type that the add-on states must be among the types the kind
declares; an outbound file must have a type that the kind declares and
must not start with an HTML page. Every transfer writes one audit line
with direction, binding, size, result, layer and time, and no name or
content. The binding page no longer says "larger files need streaming"
for a kind that streams.
The specs run the real Telegram and Slack add-ons, the real gateway and
the real file operations over scripted platform APIs, with a 6 MB file in
each direction.
Not in this change: a daily byte counter per binding, and a virus scan
(the table has a hook that accepts everything).
Assisted-by: Claude:claude-sonnet-5-5 go test
A connector add-on that declares files.transfer no longer carries file
bytes in the connector envelope. The host puts a random handle in the
request of the op media (write) or send_file (read), and the add-on
streams the file with two new host operations, connector.file.put and
connector.file.get, in chunks of at most 256 KiB.
The process that runs the add-on registers the handle when the call
arrives and ends it when the call returns. A handle is bound to the
add-on, so another add-on gets the same answer as for an unknown
handle. An unfinished inbound file is dropped with its partial bytes,
and a handle with no call for two minutes expires. Inbound bytes go
to the blob store of the workspace, outbound bytes come from it, so a
cluster needs the shared object store and no shared disk. The file is
never held in memory. The host overwrites the size and hash of an
inbound file with its own record.
The change is additive: the field files.transfer needs manifest API 2,
a kind without it keeps the inline path and its 1 MiB cap, and the host
never sends it a handle. A call that carries a handle may run up to 15
minutes. The ceiling of a file also stays under the largest file of a
workspace.
Assisted-by: Claude:claude-sonnet-5-5 go test
The grants of an agent were stored and listed, but only the proposal kinds
read them. The connector gate now reads them too, so a tool exists when
the sender holds the capability on the binding, the agent holds the grant
and the chat did not switch the tool off.
It covers triggers.propose, triggers.run, talkback.channels,
talkback.new_channel and files.upload. Every question that goes through
the gate asks the grants: the capabilities of a turn, the check of a
feature that has a sender, and a run without a sender, so a trigger that
fires from a message pauses when the grant goes.
An agent with no grant list keeps the capabilities that are not
sensitive, as before. A sensitive one (talkback.new_channel) needs the
explicit grant. Once any grant is on the list, the list is the whole
answer. The act.* grants are not enforced: the action policy of unattended
runs that uses them does not exist yet.
Assisted-by: Claude:claude-sonnet-5-5 go test
A write tool of an add-on that holds a credential was protected only by
the write mode of the connection and the confirmation of the proposal.
In a connector chat it now also needs the capability tools.write for the
sender of the turn, through the same Authorize path as the other
capabilities: the tool is not offered without it, a call is checked again
and every decision is audited. The default is no.
The rule is by shape, not by name. The manifest says an add-on holds a
credential, and contract.ToolPermission maps its non-read tools to
tools.write. The chat reads the permission from the tool definition, so
the core names no add-on.
Also close a hole in local-owner mode, where a chat turn and the browser
share one credential: a proposal that waits for a person could be
confirmed from a chat turn. The call of a chat tool now carries a mark,
and the owner refuses to confirm such a proposal under it. A remote owner
does not see the mark; local-owner mode has no remote owner.
Assisted-by: Claude:claude-sonnet-5-5 go test
One manifest-driven guard replaces the Go-only provider guard. It reads
addons/*/manifest.json and fails when the Go strings of core and pkg,
the React UI source, the locale files or the core docs hold the id of an
add-on with a role (as a whole string or a tool-name prefix) or one of
its declared titles as a whole word.
Third-party mentions (OAuth login, release sources, token formats) and
ordinary words sit in a commented allow-list. The special cases that
remain (propose_change memory kind, the Account memory tab, the Telegram
conversion, the Tables listing shim) sit in knownDebt, each with the item
that removes it, and a ceiling that only goes down.
Reword the sandbox sentences in the container docs and use neutral
sample names in the UI catalog page.
Assisted-by: Claude:claude-sonnet-5-5 go test
The provider guard flags any core string that contains the name of a
shipped add-on that provides a generic role. The web add-on has the id
and title "Web", so the built-in tools group titled "Web" in the tools
picker (and the schema example for it) failed the guard on master.
Rename the group title to "Look things up online", which matches its
"cannot look anything up online" off-state text. Update the schema
example and the Playwright fixtures and specs that read the title.
Assisted-by: Claude:claude-sonnet-5-5 go test eslint
"Do it in 10 minutes" and "at 18:00" now work from a connector chat. The
agent proposes a timer with when {"in": ...} or {"at": ...}; it runs
the model once, reports to the chat it was set in, and is removed, also
when the run failed. A timer is at least one minute and at most a year
away. Recurring triggers keep their one-hour floor. Every proposal and
confirmation text states the cost: one model run.
A proposal is answered with a plain word in the same thread: yes, ok,
no, cancel. The code stays as a fallback when several proposals wait.
Only the whole message of an allowed sender counts, and only when the
capability gate allows it, so nothing the agent read can approve.
The approval setting of a binding now takes effect. In auto mode the
trigger is made at once and the chat is told how to cancel it; in
auto_readonly mode that holds only for a proposal that says its runs
only read. The owner's limits bound it, and a proposal over a limit is
asked instead, with the reason. Recurring triggers made this way end
after the lifetime.
Assisted-by: Claude:claude-sonnet-5-5 go
A connector binding gets an approval setting: ask, auto or auto_readonly,
with four limits for the two automatic modes (live triggers, minutes
between runs of a recurring trigger, runs a day, lifetime in days). The
setting is generic, so every connector add-on can show it, and the
shared connector page of the add-ons now draws it. A binding that sets
nothing asks, as before.
This change stores, validates and exposes the setting. The next change
acts on it.
Assisted-by: Claude:claude-sonnet-5-5 go swag node
A saved agent now has a Builtins list: the capabilities its owner gave
it (memory.write, profile.write, skills.write, pages.write,
dashboard.write, triggers.propose, triggers.run, talkback.channels,
talkback.new_channel, files.upload, act.files, act.addons, act.send,
act.spend and the opt-in group saved_agents). The list is empty by
default, and nothing turns a grant on for a template, an import or a
write that omits the field.
A chat of the agent holds the grants the user may use at the turn
(agent scope). A tool exists when the grant is held, the chat did not
switch its group off and the run can hold rights: a sub-agent or a run
that started by itself never does. propose_change and revert_change
now exist only for a chat that holds the grant of at least one kind,
and the schema lists only those kinds.
propose_change gains the kinds skill, page and dashboard. A skill is a
SKILL.md for the personal or the agent layer, and a skill with
locked: true is refused, also as a shadow of a locked personal skill.
A page is a set of files checked against the Pages rules and written
as one change set. The dashboard is a home.json checked against its
schema. All three are re-checked when the owner confirms or edits.
GET /api/chats/agents/builtin-grants lists each grant with its title,
description and sensitive flag. The agent record, the write body and
the export file carry builtins; an old file imports as before and an
unknown name is dropped in an import and refused in a write.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt swag
A connection now has a write mode, chosen on the add-on page: off (the
default), propose (each change waits for the person to confirm) or auto
(a chat of the person makes it at once). The host enforces it for every
write tool of an add-on that holds a credential, before the add-on is
called, and holds the call to the repository list of the connection.
Host:
- Binding.WriteMode, stored in both stores and in the connector API
(write_mode). Absent means off, so existing bindings stay read only.
- A credentialed add-on may declare write tools. They name the resource
with the credential slot argument, never resourceId. A tool can set
approval "always", a ceiling that keeps a proposal even in auto.
- The proposal broker fetches the credential for the preview and again
for the apply, so it is never stored. A mode lowered between the
proposal and the confirmation refuses the apply. A proposal that
awaits a person cannot be confirmed by a chat turn.
- New capability name tools.write for the roles a connector kind
declares.
Add-on, tools (kind write): comment-issue, comment-pr, create-issue,
edit-issue (title, close or reopen), add-labels, review-comment-pr.
They preview plain facts, refuse a stale target, cap and screen the
text for secrets, and keep a receipt journal plus a hidden marker in the
text, so a change whose answer was lost is not sent twice.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt ginkgo node
Ettore Di Giacinto