Commit Graph
231 Commits
Author SHA1 Message Date
sam 320c990352 fix(gateway): save a messaging session instead of requiring web visibility 2026-10-20 03:12:40 +11:00
sam 5c47ebd05b feat(chat): ask several questions in one interaction with ask_questions
A turn can pose a questionnaire (single and multiple choice, free text,
skippable) and read the answers back, rendered in the chat with the
shadcn-vue questionnaire. The gateway contract gains a questionnaire ask and a
can_question capability; answers travel as a JSON object keyed by question name
and are checked against the questions that were asked.
2026-10-19 03:30:47 +11:00
sam 808551aeb0 feat(ui): Build with Archie runs on the chat system, floats over the canvas and keeps its conversation
Each workflow has its own persisted conversation, rendered like the chat panel.
The canvas and YAML go read-only while Archie works instead of inert. The code
block highlighter loads Go and TOML and drops jsx, tsx and vue.
2026-10-19 02:43:54 +11:00
sam b3e055245b fix(ui): Instructions replaces SOUL in routes and agent pages; terse alerts; missing-key hint 2026-10-18 19:27:11 +11:00
sam 53db182e97 feat(gateway): read instructions and access policies from the workflow tools
settings_get now reads three more kinds:
- soul, the instruction text layered into a workflow run;
- task-policy, the diff cap and alert delivery settings;
- org-model-policy, which models an org may use.

None of them holds a secret value: task-policy's notify_secret and the
secrets elsewhere in the readable kinds are SecretRef references (engine and
key), never the resolved value.

access_policies_list is new: the caller org's Cedar policies (id, level,
scope, text), and the instance policies when the caller owns the instance,
which is what the dashboard lists for the same read. It is guarded as
access.ActionRead on access.KindPolicy, the check the dashboard makes.
IntegrationSource gained Policies, implemented over the State Store's
ListPolicies.

workflow_get now says how its instructions field layers: after the
instance's and the organisation's, before the repository's.
2026-10-18 02:39:25 +11:00
sam c19889ac47 fix(gateway): workflow and integration tools ask the access policies as their caller 2026-10-18 02:35:59 +11:00
sam 0b12b5d387 feat(gateway): workflow and integration tools for chat and Kit harnesses alike 2026-10-18 02:32:30 +11:00
sam 1409193feb feat(workflows): iterate on a workflow's draft with Archie from the canvas 2026-10-18 02:27:41 +11:00
sam c0ae205bdf refactor(curator): rename SessionSummary to SessionRef 2026-10-17 17:12:52 +11:00
sam 94cb569f01 feat(workflow): configure review precision and approval on the review.pr step 2026-10-17 17:00:55 +11:00
sam 24ded57b4e feat(daemon): execute and verify organisation repository connections 2026-10-17 09:32:39 +11:00
sam 432cbf0846 feat(ui): give organisation, instance and account clear homes
The Access area mixed organisation, instance and personal settings under
one name. Split them: /org is the organisation's own area (members,
identities, connections, models, schedules, playbooks), /settings is
instance administration only, and personal security stays in the account
menu. Setup checklist steps now link to the page that fixes them.
2026-10-17 09:01:41 +11:00
sam ecca842bfc feat(ui): configure forge connections in the dashboard
A Connections page under Settings lists the organisation's forge connections:
host, forge type, the reference to the credential (never its value), the
external account it belongs to, and the operations each execution identity may
perform through it. The chat agent's page registry names the page too.
2026-10-17 03:20:17 +11:00
sam abb87b3b85 refactor(ui): move personal tokens and instance authentication out of Access
Personal API tokens are the signed-in operator's, and the instance URL and
OIDC providers are the instance's, so neither belongs under the organisation
Access area. Tokens render in the account page and Authentication moves to
Settings; the old paths redirect.
2026-10-17 02:41:24 +11:00
sam ff5c27ba72 refactor: remove personas
Personas were a misread requirement: SOUL is the chat identity and
identities are per action. The selected persona's prompt never reached the
model. Drops the resource, /personality, the chat selector, the settings
page and ChatService.SetPersona, and deletes stored persona documents.
2026-10-16 03:57:08 +11:00
sam 997887dada feat(gateway): chat reads workflows and validates drafts
workflow_list, workflow_get and workflow_step_types read the org's
definitions and step catalog; workflow_validate runs the editor's save
check on a proposed definition without saving it.
2026-10-16 03:46:27 +11:00
sam 2f6187d17b refactor(ui): unify account administration under Access 2026-10-16 02:57:13 +11:00
sam aa012a7c6e feat(identity): add browser setup and user accounts 2026-10-16 02:41:24 +11:00
sam a8053f3b8f refactor(gateway): delete unused wrappers and in-memory session store
Remove gateway re-exports of messaging helpers with no callers, the
unused DeltaFunc adapter, DefaultPersonas, NewStoreTaskCreator, the
SortBySeq stub, and the in-memory session store that nothing constructs.
2026-10-14 23:10:34 +11:00
sam 49233efe0a fix(gateway): isolate model selections by org 2026-10-14 21:19:34 +11:00
sam 21c0b0f0a1 fix(gateway): restore provider model selection in chat
Restore provider groups and concrete model references in chat snapshots. Permit catalog selections while retaining named aliases and tenant model restrictions. Show the active model in /profile without requiring a task identity.

Refs: archie-core-mvek6
2026-10-14 21:03:30 +11:00
sam cd983c7877 fix(webui): shorten route descriptions, setup steps, alerts and panel errors 2026-10-13 18:10:39 +11:00
sam 8a1e07f93e fix(agent): drop the one-question cap from the chat prompt and SOUL 2026-10-13 17:49:35 +11:00
sam 3ab3a5b5e1 fix(webui): attention vocabulary no longer addresses a single user 2026-10-13 17:42:11 +11:00
sam 358b964ba7 feat(webui): finish redesign dashboard task and secret settings 2026-10-13 16:41:35 +11:00
sam eba2e8e53a fix(gateway): remove the web /restart command that no process could serve 2026-10-11 04:11:37 +11:00
sam 57938158af fix(memory): each org's agent memory is its own 2026-10-10 21:20:37 +11:00
sam 3efcaf4046 feat(gateway): chat sessions belong to one org
Sessions carry an org; listing, reading and writing a session, its
messages and its turns are limited to the caller's org. Another org's
session reads as not found and cannot be overwritten or appended to.
Sessions from before orgs belong to the system org.
2026-10-10 18:57:26 +11:00
sam 87034535b6 feat(archiegateway): chat resolves models in the caller's org and bills it
The State Store answers the calling principal's org aliases
(CallerModels) and bills chat usage to that org; usage records no longer
carry a caller-named org. Chat lists and resolves aliases for the
caller's org and fails rather than substitute one it may not use.
2026-10-10 18:53:57 +11:00
sam f4e1308b23 feat(config): model aliases replace model roles, resolved by one resolver
Agent steps, implement, pr-review, MCP sampling, chat, embedding and
transcription resolve a model alias (empty means the purpose's default)
and fail on an unknown alias instead of falling back. The step editor and
chat pick aliases; workflow definitions naming an unknown alias are
refused on save.
2026-10-10 17:29:52 +11:00
sam 80f6462bc4 refactor(gateway): drop duplicate string arg helpers 2026-10-10 11:05:59 +11:00
sam 66948c2deb fix(gateway): attribute chat settings to signed-in users 2026-10-09 09:57:57 +11:00
sam 8f7c14e5c6 fix(gateway): read task logs through state store 2026-10-09 09:57:57 +11:00
sam cc64faa585 feat(chat): the dashboard chat records voice, downloads files a turn sends, and offers /model and /personality choices 2026-10-08 01:37:21 +11:00
sam b0f1ad7cdc feat(telegram): /stop stops the chat agent, else the tasks it created, or one task; delete the dead approval actions 2026-10-06 01:12:28 +11:00
sam 151f847e21 refactor(gateway)!: drop budget_key with the webhook chat channel 2026-10-05 18:10:41 +11:00
sam 23307d9462 feat(gateway): load SOUL into the chat prompt per turn
SOUL now owns the prompt identity slot via a per-turn source: a stored soul
resource wins once saved, otherwise the identity's SOUL file, then the
shipped default. The gateway watches the resource so a dashboard edit
reaches the next turn without a restart; a file read error degrades to the
default and logs once.

The persona prompt plumbing is replaced by the SOUL source; the persona
surface itself (registry, wire, page, commands) stays for archie-core-7ftt.
2026-10-05 17:14:20 +11:00
sam 24374f026b fix(webui): bind the container runtime document's keys, drop the blank Advanced page
ContainerRuntimePage still bound the Go field names the document carried
before archie-core-qna6 gave container-runtime-policies its own snake_case
shape. Every control read a key the document does not hold, so the fields
rendered empty and DurationInput warned on each mount; the page's Profiles
section wrote to a key the decoder drops. Bind image, max_concurrency,
max_uptime, volume_ttl, pull_policy and network, drop the dead Profiles
section (named profiles live in the agent-profiles resource, filed as
archie-core-x0wk), and pin the key list in a test the migration lacked.

Settings -> Advanced rendered nothing in any deployment: its only card needs
the sandbox dangerous-command authority, which has no implementation
(archie-core-sod1). Drop it from the sidebar, the router and the gateway's
dashboard page index; the page and its components stay in the tree for sod1.

Register CardAction in SetupPanel and EmptyDescription in TaskLogs, which Vue
was resolving to nothing, so the setup percentage and the log empty-state
descriptions render again.
2026-10-04 20:51:57 +11:00
sam 96a90d643b refactor(gateway): cut comments down to what the code does 2026-10-03 21:53:24 +10:00
sam 3c338a9722 refactor: strip doc and issue narration from code comments
Remove PRD/doc paths, bead IDs, phase notes and deleted-test references
from comments; drop test-only hooks left by the test deletion. Includes
in-progress archied bootstrap, gateway and query changes.
2026-10-03 21:47:46 +10:00
sam a45300d626 refactor: testing from the beginning. a journey in 150k lines of tests to none 2026-10-03 21:16:25 +10:00
sam 6fea86ae5e feat(gateway): offer ask_user when the turn's channel can carry a question
prepareTurn now registers an ask_user tool only when the turn context
carries an interactive adapter (WithInteractive), so a contract with no
caller has a real one and a channel that cannot ask registers no tool
rather than one that always fails. A channel reporting Picker gets the
native single-choice request; one reporting only Clarify still carries a
choice, rendered as numbered text and matched from the typed reply through
the domain's FormatPickerText/ParsePickerReply rather than erroring.

Red-first: InteractiveTools returned nothing and
TestInteractiveToolsRegisteredWithCarrier / TestAskUserUsesPickerWhenChannelCanPick
/ TestAskUserFallsBackToTextWhenChannelCannotPick failed at the assertion
before the tool was implemented. TestTurnRegistersAskUserOnlyWithCarrier
proves reachability from a real TurnRunner turn.
2026-10-02 11:52:07 +10:00
sam a8ea4e0656 fix(messaging): transcribe voice notes on the model-owning side of the boundary
The Messaging Service built the voice-transcription client in-process
from [models].transcription and [providers.*], which the messaging
boundary PRD reserves to the process that owns the model runtime and its
provider credentials. The deletion gate only measured linked packages,
so the PRD violation shipped green: config reads are not imports.

Move the capability across the boundary:

- archiemessaging no longer reads [models] or [providers], builds no
  transcription client, and drops the ResolvedConfig.Transcriber field.
- telegram classifies a voice note as messaging.MediaTypeVoice and stops
  transcribing; it carries the attachment bytes and its note to the turn,
  which is all a frontend without the credential can do.
- archied builds the transcriber beside the chat runtime from its own
  model-role configuration (setupTranscriber) and gives it to every
  gateway TurnRunner.
- the turn runner transcribes a speech attachment before recording the
  inbound message, so stored history and the model context carry the
  transcript; a nil/failed/empty transcription keeps the frontend note.
- messaging.TranscribedMessageText renders the transcript into the text a
  frontend built from its media note, preserving the sender's caption.

Red first: TestTurnRunnerTranscribesSpeechOnTheModelOwningSide failed at
'Transcribe calls = 0, want 1' with the transcription call disabled.
2026-10-02 11:12:59 +10:00
sam 3a65fd2ee1 feat(controlplane): add the writable review-settings resource
review.precision_gate and review.approve_before_post were file-only, so the
dashboard had no way to write them: no control-plane resource carried them.
Add the review-settings kind (docs/prds/review-settings-resource.md), seed it
from the file's [review] section, layer it field-by-field into cfg.Review,
watch it live, and publish it through ConfigView.Review. The dashboard gets a
Review settings page bound to the kind through PAGE_RESOURCES, showing each
dial beside the value in force.
2026-10-02 11:08:44 +10:00
sam 00e5b431e4 feat(webui): let the operator choose the retry worktree mode
A retry now carries the operator's worktree choice (refresh onto the base
branch or continue the work already pushed on the task branch) from the
dashboard through the task action to the daemon dispatch, and persists it
on the task row so it survives a reload. The daemon resolves the mode from
the task, never from the workflow; BeginRemediation records continue
explicitly so a remediation still resumes its pull request branch.
2026-10-02 10:53:21 +10:00
sam 4192a9f27c fix(archied): omit the max-token bound on reasoning-class models
The OpenAI chat-completions provider in the pinned ai-sdk serialises a
non-zero output-token bound as `max_tokens`, which reasoning-class models
reject with HTTP 400 ("Use 'max_completion_tokens' instead"). Send no bound
at all for a model the catalog marks reasoning; classic models keep their
bound. Covers the chat turn path and MCP sampling, the two call sites that
set one today.

The reasoning flag rides TurnPrepareContext from the catalog details the
gateway already resolves, so archie-core keeps one notion of "reasoning
model": messaging.ModelDetails.Reasoning. A max_completion_tokens path in
ai-sdk is still needed to ever bound these models; that is filed separately.
2026-10-02 10:25:39 +10:00
sam e6f90b99e8 feat(gateway): gate admin slash commands by sender policy
Add SlashAccessPolicy, a deny-by-default role policy (admin allowlist plus
per-user delegated commands), and enforce it in Router.route before any
local command dispatches. A refusal executes none of the handler, is
visible to the sender, logged, and recorded through an optional
SlashDenials sink.

No config field yet: the gatewayrpc wire drops Message.SenderID (only the
display Sender crosses), so a per-sender allowlist cannot match a Telegram
sender at the daemon Router until the layered auth mixin carries identity
(archie-core-263). Nil policy preserves today's behaviour.
2026-10-02 10:00:37 +10:00
sam acfba9121f feat(messaging): adapter-facing clarify, picker and approval contract
The optional-capability pattern had media and delete but no interactive
UX: a channel could ask a human to approve an action through
ApprovalRequester, but there was no domain contract for posing a
clarifying question or presenting a single-choice picker, and no
capability flag a caller could read before attempting one.

Add ClarifyRequester and PickerRequester to internal/domain/messaging
alongside ApprovalRequester, with Clarify/Picker/Approval flags on
AdapterCapabilities and ClarifierOf/PickerOf/ApproverOf lookups that
treat the capability as the contract rather than the method set, so a
mis-reporting adapter degrades instead of panicking.

TextFallback is the shared text degrade: a channel with no native
question or selection UI still carries the interaction (numbered
prompt, parsed reply) instead of failing or silently dropping it.
FormatClarifyText/FormatPickerText/ParsePickerReply hold the rendering
and matching logic once.

The approval half reuses the existing decision model: Telegram's
telegramApprover already renders a native prompt, so it now reports the
Approval capability and deliberately does not claim Clarify/Picker it
has no native UI for. Gateway re-exports the contract.

Red-first: internal/domain/messaging/interactive_test.go drives the
test double adapters (native, degrade, mis-reporting, over-claiming)
and the text rendering/parsing; telegram/approval_test.go pins the
honest capability report.
2026-10-02 09:57:44 +10:00
sam 2d200de224 feat(gateway): list the harness page in the dashboard page registry
internal/gateway/dashboard_tools.go mirrors the dashboard's route table so
the agent can navigate to every page. The new /settings/harness route needs
its entry or page_index would not offer it (archie-core-egkf.7).
2026-10-02 09:38:09 +10:00
sam 15946dacfb feat(taskactions): one approval path answers the review gate from every surface
Decision 1 of docs/prds/pr-review-operator-response.md: the dashboard's
POST /api/tasks/{id}/action is the authoritative surface, chat's /approve
is a shortcut that posts every offered finding, and both reach the same
Service.Apply so one operator intent cannot be recorded as two decisions.

The action body now carries instructions (required by rereview) and the
finding keys an approve posts. taskactions.Service answers an offer
through the guarded RespondReviewGate write: approve records the
selection on the offer, rereview records the instructions, increments the
round and clears the offer, and a rereview with no instructions, with no
offer, or at the cap is refused as a conflict with nothing written. Chat
reaches the same service: chatTaskControllerAdapter.ApproveChatTask
delegates to b.taskActions() instead of its own requeue, which is what
removed the handler's hardcoded implement handoff -- feasibility's deliver
stage now names implement on the task before it waits, the invariant the
handler needs.

The payload crosses every hop it must: the NATS action request, the
gateway.v1 gRPC contract (regenerated), the LocalChatAdapter, and the
webui handler, which answers 400 for a rereview without instructions and
409 for the cap. A rereview event kind names its actor the way the
approval and rejection trios do.

taskactions.Task gains review_gate and rereview_rounds, and the two
surfaces' tests assert they agree.
2026-10-02 02:27:59 +10:00