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
StripPathPrefix and the redirect from / and /browse still read
X-Forwarded-Prefix from any peer, after BaseURL and BasePathPrefix
were gated on --trusted-proxies. A client could make the server strip
a path prefix or add one to a redirect.
Gate both on the same trusted-proxy resolver. The resolver is now
stamped into the request before StripPathPrefix runs, and the check is
exported as ForwardedHeadersTrusted. The explorer has no
--trusted-proxies option and keeps its old behavior by trusting every
peer for this one header. The docs say that the prefix needs a trusted
proxy.
Assisted-by: Claude:claude-sonnet-5-5 go-test
Describe the install step, held upgrades, the Host access page, deny
rules and the blanket switch.
Assisted-by: Claude:claude-sonnet-5-5 playwright eslint
A GitHub write that waits for the person used to show the raw preview
JSON in the host's approval frame. The add-on now declares an
approval-card component for each of its six write tools. It shows what
will be done, where (with a link to the issue or pull request), as which
connection, and the exact stored text that will be posted, plus a note
that the text is public.
The proposal does not name the connection, so the card reads the saved
connections and names the one that lists the repository.
Assisted-by: Claude:claude-sonnet-5-5 node playwright eslint
Token help now works: a button opens GitHub's fine-grained token page,
the steps are open while there is no connection, and a table lists the
permissions for read only and for write modes. The page says it does not
answer mentions yet.
Errors sit under their own field. The repository list splits on lines and
commas only, so "not a repo" is one entry and is named in the message.
The connection name defaults to the first repository.
The write mode cards each carry one consequence line, and "Allowed" is
now "Acts without asking". The Test result is drawn per repository in the
row, with a tone, a busy line, and a plain text when GitHub refuses the
token. The unit tests and a browser spec cover the page.
Assisted-by: Claude:claude-sonnet-5-5 node playwright eslint
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
Describe schemaVersion 2 and its deny object, the deny entry grammar, the
--addon-host-deny flag, the blanket and UI grant settings, and the rule
that a selected policy file is always opened for its deny, even when a
flat --addon-host-ops list replaces its grants.
Assisted-by: Claude:claude-sonnet-5-5
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
BaseURL and BasePathPrefix read X-Forwarded-Proto, X-Forwarded-Host,
X-Forwarded-Prefix and Forwarded from any caller when LOCALAI_BASE_URL
is not set. A client could pick the scheme and host that end up in the
SPA <base href>, generated links and redirects.
Gate them on the trusted-proxy resolver (--trusted-proxies): the headers
count only when the TCP peer is trusted. An empty list trusts nobody, so
the request's own scheme and host are used. LOCALAI_BASE_URL still wins
over everything for the origin.
This changes behavior for setups behind a proxy that rely on these
headers without setting LOCALAI_BASE_URL: they must now list the proxy
in --trusted-proxies. The docs say so.
Assisted-by: Claude:claude-sonnet-5-5 go-test
Format the HTTP route constants with gofmt.
Assisted-by: Codex:gpt-6
(cherry picked from commit 9d4b99da60)
(cherry picked from commit 367a9ff9432c84f700f3f83206fab2d7b4ed85ac)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The speech-to-speech path labels its transcription WAV with the session
rate for client commits, so resampling first handed it 16 kHz audio under
a 24 kHz header. Resample only in the chained pipeline, and say so in the
docs.
Assisted-by: Claude:claude-sonnet-5-5 go-test
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
(cherry picked from commit 82cc411af54cf884e4a44a3e96292a323ba5611c)
The committed-utterance WAV is written with a fixed 16 kHz header. The VAD
commit path stores audio that is already resampled to 16 kHz (vadCommit),
but the client-commit path (input_audio_buffer.commit) passed the raw
buffer at the session input rate. For a 24 kHz OpenAI Realtime session the
backend decoded 24 kHz audio as 16 kHz (1.5x speed/pitch): garbage or empty
transcripts, then input_not_committed on the follow-up response.
commitUtterance now resamples from session.InputSampleRate to
localSampleRate, so both paths hand 16 kHz audio to the 16 kHz header.
This replaces the earlier approach of this PR (header at the session rate),
which mislabelled the already-resampled VAD audio (review feedback).
Tests go through the real paths with a 24 kHz session: commitUtterance and
a server_vad commit via vadTick; both must produce a 16 kHz WAV of the
resampled length. The client-commit spec fails without the fix.
Fixes#12747
Signed-off-by: nexxtmobile.de <git@nexxtmobile.de>
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
(cherry picked from commit 05ba891da6)
(cherry picked from commit 56e2b6d0cc5cbf1944ef50735f7b499d998eb25c)
Clients written for llama-server, vLLM or SGLang switch template variables
per request with a top-level chat_template_kwargs object, most often
{"enable_thinking": false}. OpenAIRequest had no such field, so it was
dropped: with a Qwen3-style template thinking stayed on, the reply landed
in message.reasoning and content came back empty.
Fold the field's scalar values into the request metadata, so it takes the
path metadata already takes (gRPCPredictOpts, ResolveChatTemplateKwargs).
Booleans become "true"/"false", strings pass through, numbers become their
decimal string; nested values are skipped. An explicit metadata key wins.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: Cristi An <c4xp@msn.com>
(cherry picked from commit 2cd7425e87)
(cherry picked from commit 5e86ad6bd5286cc8db398d278adaa32e562223ca)
Drop the vendor marketing claims and the specific upstream model ids,
which go stale. Keep the working passthrough configuration.
Assisted-by: Claude:claude-sonnet-5-5
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
(cherry picked from commit eec17103b4514f1cb0f6193c3683f34f71a3d1b9)
Document OrcaRouter as a first-class cloud-proxy upstream. It is an
OpenAI-compatible gateway, so it plugs into the existing passthrough
mode with provider: openai, mirroring the OpenAI/Anthropic examples.
Assisted-by: Claude:deepseek-v4-flash-0731
Signed-off-by: XiaoHuo888-hue <xiaohuo888hue@users.noreply.github.com>
(cherry picked from commit 329ee0ee89)
(cherry picked from commit 6b2d84294d0198bdaa2f27288993eb7cfad0b0b1)
Report failed warmup requests without failing model load. Document the
warmup limits and use the reported LocalAI benchmark measurements.
Assisted-by: Codex:gpt-6
(cherry picked from commit ec50871dd6)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
A model can cite a plausible deep link it never opened, written from
memory. The client cannot tell it from a link a search tool returned,
because server-side MCP tool results never reach it.
evidence_check.urls compares every http(s) link of the final answer,
bare or inside a Markdown link, with the URLs of the request (system
prompt, system, user and tool messages) and of the request's successful
MCP tool results. Links are compared without scheme, fragment, "www."
and trailing slash; a link counts as retrieved when it equals an evidence
URL or is a shorter form of one at a path or query boundary. Unbacked
links (at most five) are listed in an appended note; urls_note replaces
the default note, with {urls} as the placeholder for the list.
Assistant messages are not evidence, so an invented link is marked again
when the client sends it back on the next turn.
Assisted-by: Claude Code:claude-opus-5-5
Signed-off-by: Stefan Walcz <stefan.walcz@walcz.de>
(cherry picked from commit 523e8cd3c4)
Keep balanced parentheses in source URLs and decode MCP text arrays before
extracting links. This preserves Wikipedia paths and escaped query
separators instead of appending truncated citations.
Assisted-by: Codex:gpt-6
Signed-off-by: Stefan Walcz <stefan.walcz@walcz.de>
(cherry picked from commit 9b81cb20ab)
A system prompt that tells a model to look things up first does not stop
it from answering from its own knowledge, for example when the question
comes with a picture.
profile.require_tool makes the first model call of a chat request that
offers the profile's server-side MCP tools use tool_choice "required".
Every follow-up call of the server-side tool loop is made with the
client's tool_choice again: a "required" that stayed in place would make
the model call a tool after every tool result and never answer until the
iteration limit.
With a tokenizer template the backend enforces "required" with a grammar
built from the chat template's tool-call format. Where LocalAI builds the
tool-call grammar itself, the first call's grammar leaves out the
no-action function instead.
The first call is not forced when no server-side tool was offered, when
the client set tool_choice or function_call itself, or when the request
does not end with a user message, so a request returning a client tool's
result cannot bounce tool calls between client and server.
Assisted-by: Claude Code:claude-opus-5-5
Signed-off-by: Stefan Walcz <stefan.walcz@walcz.de>
(cherry picked from commit 2801938aac)
A research profile asks the model in its system prompt to cite its
sources, but a model can answer from a search result without naming a
single link, and the client never sees the server-side tool results that
held them.
evidence_check.sources appends one line with the distinct http(s) URLs
returned by the request's successful MCP tool calls (at most five, in
order of appearance) when the final answer names no URL itself. An answer
that already links something is left unchanged, and nothing is added when
no tool result holds a URL, so a link is never made up. The label is
configurable through evidence_check.sources_label (default "Sources") for
profiles that answer in another language.
It runs in the existing evidence gate, on the same streaming and
non-streaming paths as the number note.
Assisted-by: Claude Code:claude-opus-5-5
Signed-off-by: Stefan Walcz <stefan.walcz@walcz.de>
(cherry picked from commit 4ee6ff1395)
Reserve capacity when selecting a worker and release it when forwarding
ends. Completed connections no longer bias routing toward new workers.
Keep active counts through offline periods and choose randomly on ties.
Add regression tests and document per-instance TCP connection counts.
Fixes#12992
Assisted-by: Codex:gpt-6
(cherry picked from commit 66cf106490)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
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 Tools list gains a group, "What the agent may change", with one
row per builtin grant: its title and a plain-words description, and a
mark on the sensitive ones. The agent form switches them through
useAgentTools; a chat of a saved agent lists the grants the agent
holds, read only. Both draw the group with the same useToolGroups and
SwitchPanel code.
The group has no group setter, so Select all and Disable all, and the
"N of M on" count, leave it alone: a bulk click must not grant a
right.
Assisted-by: Claude:claude-sonnet-5-5 node
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 add-on guide names one manifest api, one way to move a file and no
migration path, and drops the section on arguments of older tool
versions. The Swagger files are generated again, which also brings in
fields that the last generation missed, and the test of the apps view
now emits every field it documents. The inventory of the add-on suites
lists the specs by their new names.
Assisted-by: Claude:claude-sonnet-5-5 go swag
No release ever shipped add-ons, so there is no installed base on api 1.
The manifest must now declare api {"min": 2, "max": 2}; a manifest with
api 1, no api or a wider range is refused with a message that names the
range.
Every field that was gated on api 2 is now part of the contract, so the
Min*API constants and their gates are gone, and so are HostAPIMin and
the apiMin host field. Three older-manifest shims go with them: the
resource listing the host implied for one add-on, the legacyArgs hint
for arguments an older version of a tool took, and the tests that
checked the gates. All shipped manifests and the test fixtures declare
api 2.
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
Echo's RealIP trusted X-Forwarded-For and X-Real-Ip from any peer
when no IPExtractor is set. That fed the login and registration rate
limits and the API trace client address, so a client could dodge a
limit by rotating a forged header.
Add pkg/clientip, one resolver for the whole server. With
--trusted-proxies / LOCALAI_TRUSTED_PROXIES (IPs and CIDRs, empty by
default), the forwarded chain is read only when the TCP peer is
listed. The client is the right-most X-Forwarded-For entry that is
not a trusted proxy; a malformed entry stops the walk. With the flag
unset every request resolves to its TCP peer.
The resolver is set as Echo's IPExtractor and used by the access log
(--log-client), so all consumers agree.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt
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
Bindings saved before permissions were enforced have no capabilities, so
they deny files from senders, triggers, memory writes and talk-back, and
the owner could only fix that through the REST API.
The shared connector page now has a section for it, so every connector
add-on gets it: a switch and a description for each capability with the
sensitive ones marked, the role matrix (never a sensitive capability, never
above the ceiling), pairing with a code shown once, the approved channels
for talk-back, and a read-only view of the access check with the latest
decisions. A binding with no capability shows a notice that says why the
features are off, with an "Allow the usual" button that turns on the
non-sensitive everyday set.
The kinds endpoint returns the closed list of capabilities with a title,
a description and the sensitive and usual flags. The field is additive.
Assisted-by: Claude:claude-sonnet-5-5 go node playwright swag
"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 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
The four document skill names were code in the contract, the hint for
binary files kept its own list of formats and libraries, and the chat
boundary rewrote the old arguments of one add-on by its ID.
A skill that shadows a name can now declare the slot it serves: kind,
file extensions, name words and a library hint. A tool can declare
legacyArgs, the arguments it refuses now, each with a hint. Both need
manifest api 2. The four default slots move to a data file that the
contract reads, so a manifest that declares nothing keeps working.
The host takes the replaceable names, the files a skill covers and the
hint for a script from the skills the add-ons list. It compares no
format or add-on ID. A guard spec fails when it does again. The Office
manifest declares its slots and legacy arguments and accepts api 2.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt
A connector chat whose sender holds talkback.channels gets the tool
send_message. It writes in the conversation, or as a new top-level
message in a channel that the owner listed on the binding (id, label,
approved; at most 10). The tool names no platform.
An unlisted channel is refused and the error lists the ones that are.
A listed channel that the owner did not approve needs the sensitive
talkback.new_channel for each message, so only a paired sender can use
it. The agent cannot add or approve a channel, so text it read in a page
or a file cannot widen where it writes.
Messages go through the outbox (rate limit and queue), with the size cap
and the mention breaking. Each attempt adds an audit row with channel,
kind, capability, length and a hash, never the text.
A trigger can report to an approved channel with delivery target
channel. The binding, the capability and the listing are checked again
when the run ends; any failure sends the notice to the Inbox.
LOCALAI_DISABLE_CONNECTOR_TALKBACK turns both off.
Assisted-by: Claude:claude-sonnet-5-5 go swag
Document reasoning.effort and request metadata forwarding for the Responses API.
Assisted-by: Codex:gpt-6
(cherry picked from commit e86876c6f3)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
Anthropic sends tool_choice as an object: {"type": "auto" | "any" |
"none"} or {"type": "tool", "name": ...}. The endpoint only read the
"tool" object, so {"type": "none"} still offered the tools and
{"type": "any"} did not require a call. The object was also forwarded
to the backend unchanged, and the llama.cpp backend turns any object
tool_choice into "required", so even {"type": "auto"} forced a tool
call there. Read the object form, and forward the OpenAI equivalent.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: jaeminbird <jaeminbird@gmail.com>
(cherry picked from commit 5bd07865a3)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
/v1/messages always answered stop_reason "end_turn" for text replies,
even when generation stopped at max_tokens. Clients that continue a
truncated turn (the Anthropic SDKs, Claude Code) treated the cut-off
reply as complete. Use the same check the OpenAI endpoint uses for
finish_reason "length", in both the streaming and non-streaming paths.
Assisted-by: Claude:claude-opus-5-5
Signed-off-by: jaeminbird <jaeminbird@gmail.com>
(cherry picked from commit d905ab5fdc)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The mutable upstream download no longer matches the gallery checksum.
Pin the file to an immutable revision with the published expected hash.
Regenerate the compatibility gallery and document installation.
Fixes#12924
Assisted-by: Codex:gpt-6
(cherry picked from commit 800855297b)
Signed-off-by: Ettore Di Giacinto <mudler@localai.io>
The finetune.start and quantize.start operations defaulted to the
backend names "trl" and "llama-cpp-quantization", and the backend
lister had one method per engine family. Both were special cases that
a new engine could not use.
The lister now answers for a gallery tag. A start call that names no
backend gets the first installed backend of the tag in gallery order,
and a named backend must carry the tag. With no engine installed the
call fails with a 400 that says so. The shipped gallery order keeps
trl and llama-cpp-quantization as the defaults, and a test pins it.
The change is additive for add-ons: no manifest field or operation
changes.
Assisted-by: Claude:claude-sonnet-5-5 go gofmt
Ettore Di Giacinto