Files
LinYushen cb68669c73 feat(composio): gate MCP apps behind feature flag (#4876)
* feat(composio): server-side connect flow + connections REST (Notion MVP) (MUL-3720) (#4608)

* feat(composio): server-side connect flow + connections REST (Notion MVP) (MUL-3720)

Compose the merged server/pkg/composio SDK into a user-facing connection
manager: signed-state connect handshake, local user_composio_connection
mirror, idempotent disconnect, and a per-user MCP session helper (not yet
wired into task dispatch).

- migration 127_user_composio_connection (no FK/cascade, per DB rules)
- sqlc queries: upsert (idempotent on user_id+connected_account_id), list
  active, owner-scoped get, mark revoked
- internal/integrations/composio: signed HMAC-SHA256 state, BeginConnect,
  CompleteCallback (idempotent upsert), ListConnections, Disconnect
  (upstream 404 = idempotent success), CreateMCPSession (no-op when empty,
  pins connected_accounts per toolkit), CallbackRedirect
- REST handlers under /api/integrations/composio (user-scoped, 503 when
  COMPOSIO_API_KEY unset): connect/init, callback (302), connections list,
  delete
- router wiring gated by COMPOSIO_API_KEY; COMPOSIO_AUTH_CONFIGS_JSON maps
  toolkit->auth_config (MVP: notion); state secret from COMPOSIO_STATE_SECRET
  or derived from JWT_SECRET; callback base from COMPOSIO_CALLBACK_BASE_URL
  or MULTICA_PUBLIC_URL
- tests: state (expire/tamper/wrong-secret), service (mapping, callback
  idempotency, non-success, disconnect owner/404 idempotency, MCP pin),
  handlers (httptest), redact regression for Bearer mcp_ tokens

MVP scope: Notion only; no task-dispatch overlay, sharing, or webhook
event handling (later stages).

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): bind callback account to user + idempotent revoked disconnect (MUL-3720)

Address PR 4608 review (CHANGES_REQUESTED):

- callback: verify connected_account_id with Composio before mirroring it.
  The signed state only proved user/toolkit/exp, so a valid state paired with
  a tampered connected_account_id would be written verbatim. CompleteCallback
  now calls ListConnectedAccounts and fails closed (ErrAccountVerification)
  unless the account belongs to the state's user (composio_user_id == multica
  user id) and was created under the toolkit's auth config. No row is written
  on mismatch / unknown account / upstream error.

- disconnect: short-circuit to a no-op when the local row is already revoked,
  before touching upstream. Previously a second DELETE re-hit Composio and a
  non-404 upstream error surfaced as a 502, breaking the 204-idempotent
  contract.

- CreateMCPSession: document the v1 single-active-connection-per-(user,toolkit)
  constraint and make duplicate selection deterministic (newest-wins, rows are
  connected_at DESC) instead of order-dependent map overwrite. Stage 3 owns the
  real single-account-enforcement vs multi-account-shape decision.

Tests: tampered/wrong-auth-config/unknown-account callback rejection, revoked-row
disconnect no-op (asserts upstream not re-hit). composio pkg 85% coverage; all
green.

Co-authored-by: multica-agent <github@multica.ai>

* feat(composio): list all toolkits + dynamic auth-config resolution (MUL-3720)

Yushen's follow-up to the Notion MVP: surface the full Composio toolkit
catalog, render it in Settings, and drop the static env mapping in favor of
dynamic auth-config discovery.

Config correctness (per Composio docs):
- Remove COMPOSIO_AUTH_CONFIGS_JSON entirely. The toolkit→auth_config mapping
  is now resolved at request time from the project's /auth_configs (cached,
  5-min TTL), so enabling a toolkit is a dashboard action, not a redeploy.
- Do NOT add COMPOSIO_PROJECT_ID. The project API key (x-api-key) authenticates
  to exactly one project; the project is resolved from the key. Only org-level
  endpoints use x-org-api-key, which this integration never calls.

Backend:
- SDK: server/pkg/composio/auth_configs.go — ListAuthConfigs (toolkit_slug,
  is_composio_managed, show_disabled, limit, cursor).
- service: dynamic resolver (authConfigMap cache; betterAuthConfig prefers a
  custom/white-label config over Composio-managed, newest wins); BeginConnect
  and CompleteCallback resolve via it; ListToolkits fetches the full catalog
  (paginated, capped) annotated with connectable = has an enabled auth config,
  connectable-first ordering.
- handler + route: GET /api/integrations/composio/toolkits (user-scoped, 503
  when COMPOSIO_API_KEY unset) returning slug/name/logo/category/connectable.

Frontend:
- core: ComposioToolkit/ComposioConnection types, api client methods, and
  composio query options (@multica/core/composio).
- views: Settings → Integrations now has a Composio section rendering every
  toolkit as a card with search. Connect is gated on `connectable`;
  non-connectable toolkits show a muted "not configured" hint instead of a
  dead button. Connected toolkits show a badge + Disconnect (with confirm).
- i18n: composio block added to en/zh-Hans/ja/ko settings.

Tests: SDK + service (dynamic resolution, custom-over-managed preference,
connectable flag, resolver-error soft-degrade) and handler toolkits endpoint;
composio pkg 85.7% coverage. go build/vet/gofmt clean; core+views typecheck,
core+views lint, and core tests (691) all green.

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): close cross-toolkit callback fail-open by signing auth_config_id into state (MUL-3720)

Re-review blocker: CompleteCallback resolved the toolkit's auth config at
callback time and ignored a resolve error/empty result, while
verifyAccountOwnership skipped the auth-config comparison when the expected
value was empty. A user could then pass another toolkit's connected_account_id
into this toolkit's callback — the owner check passed and it was written under
the wrong toolkit_slug/account binding.

Fix: the auth_config_id is already resolved in BeginConnect (before the state
is signed), so sign it into the state and compare it exactly at callback. No
re-resolve, no fail-open. verifyAccountOwnership now fails closed when the
expected auth config is empty (rejects instead of skipping) and requires an
exact match — closing the cross-toolkit binding gap.

Tests: state round-trips auth_config_id; BeginConnect signs it; callback
rejects wrong/cross-toolkit auth config and an empty (no-mapping) auth config
fails closed. composio pkg 85.2% coverage, all green.

Frontend (non-blocking): the Composio settings tab now surfaces an error when
the connections query fails instead of silently rendering everything as
unconnected.

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): hide Settings section entirely when integration unconfigured (MUL-3720)

Decision (option 2, hide-then-merge): don't show a card that leaks the internal
COMPOSIO_API_KEY env-var name to every end user. IntegrationsTab now gates the
whole Composio section (heading + body) on the toolkits query — a 503 means the
key is unset, so the section is withheld instead of rendering the not-configured
card. Admin-only setup guidance is a later, role-gated affordance.

Removed the notConfigured card (and now-unused ApiError import) from
ComposioTab; it only mounts when configured. views typecheck + lint clean.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>

* feat(composio): Stage 2 frontend polish — callback toast, last_used & expired UI, e2e (MUL-3718) (#4688)

* feat(composio): callback toast + refresh, last_used & expired UI, e2e (MUL-3718)

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): real callback redirect route + StrictMode-safe toast dedup (MUL-3718 review)

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): callback endpoint should not require Multica auth (MUL-3843) (#4709)

* fix(composio): move OAuth callback out of the Auth group (MUL-3843)

Composio 302-redirects the browser to /api/integrations/composio/callback
at the end of the OAuth flow, but PR #4608 mounted it inside the cookie-auth
middleware group. When the session cookie is absent (expired session,
SameSite=Strict / Safari ITP, private window, self-hosted callback subdomain)
the Auth middleware returned a hard 401 and a JSON blob instead of the
settings redirect, breaking the flow.

Identity never came from the cookie anyway: it is carried by the HMAC-signed
state param that CompleteCallback verifies (signature, expiry, replay) and
cross-checked by verifyAccountOwnership; h.Composio == nil still 503s. So the
callback is registered alongside the other public OAuth/webhook routes; the
other four composio endpoints stay session-gated.

Refs MUL-3843, MUL-3715.

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): correct stale callback routing comments (MUL-3843)

The package header and ComposioCallback doc comments still described the
callback as sitting under the Auth middleware group. After the route was
moved out (this PR), update both to state it is a public route whose identity
comes from the signed state — addressing review nit from 张大彪.

Refs MUL-3843.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>

* feat(composio): inject MCP overlay into agent runtime at task dispatch (MUL-3721) (#4704)

Stage 3 of the Composio epic. Wires the per-user Composio MCP session into
every agent task so the agent process sees the initiator's connected tools
without any prompt-time plumbing.

Server side
  - Migration 128 adds agent_task_queue.runtime_mcp_overlay JSONB plus a
    BEFORE-UPDATE trigger that wipes the column on any transition into a
    terminal status (completed / failed / cancelled). A trigger is the single
    source of truth — future queries that flip status cannot bypass it.
  - composio.Service.BuildTaskOverlay(userID) reuses CreateMCPSession and
    emits the Claude-style { mcpServers: { composio: { type: http, url,
    headers } } } shape the daemon's existing sidecar generators consume.
    Returns (nil, nil) on zero active connections so we never burn a
    Composio session for a user with nothing to call.
  - TaskService grows a Composio ComposioOverlayBuilder seam, wired in
    router.go after composiointeg.NewService succeeds. Five enqueue paths
    (issue / mention / quick-create / chat / auto-retry) attach the overlay
    after CreateAgentTask returns and before the daemon is notified — so
    every claim reads a settled row, with no second daemon hop. Best-effort:
    a builder failure logs and proceeds with no overlay.
  - resolveInitiatorFromTriggerComment derives the initiator user from the
    trigger comment when it was authored by a member. Agent-authored
    triggers are not treated as initiators (their connected-apps view is
    empty by construction).

Daemon side
  - handler/daemon.go claim path merges task.runtime_mcp_overlay onto
    agent.mcp_config via mergeMCPOverlay before populating
    TaskAgentData.McpConfig. Overlay wins on server-name collisions
    because it carries the live user-scoped session URL. Errors fall back
    to the agent config unchanged — a bad overlay must not surprise-disable
    saved MCP tools. The existing execenv sidecar generators (cursor /
    codex / openclaw / opencode / hermes-kiro) need no changes: they keep
    consuming the merged result through TaskAgentData.McpConfig.

Tests
  - 9 merge cases (mcp_overlay_test): both-nil short-circuit, agent-only
    pass-through, overlay-only canonicalization, two-side merge, name
    collision (overlay wins), top-level key preservation, malformed agent
    fallback, malformed overlay fallback, non-object server rejection.
  - 4 dispatch cases (composio): zero-connections returns nil without
    CreateSession, happy-path emits the right shape with the right user
    id, empty-URL defensive branch, SDK error surfacing.
  - 4 TaskService helper cases: nil Composio is a no-op (Queries-safe),
    invalid initiator does not call the builder, nil overlay skips the
    UPDATE, builder error swallowed without panic.
  - Migration 128 verified to roll up + down + up cleanly against the test
    database.

Out of scope (deferred): assignment-triggered enqueue paths with no
trigger comment get no overlay attached today (no initiator UUID flows
through enqueueIssueTask in that case). Retry paths recompute the overlay
fresh from the parent's initiator_user_id instead of inheriting the bearer
from the parent row, so a stale token can never resurface on a retry.

Co-authored-by: Eve <eve@multica.ai>
Co-authored-by: multica-agent <github@multica.ai>

* feat(composio): per-agent allowlist + originator-scoped MCP overlay (MUL-3869) (#4736)

* feat(composio): per-agent allowlist + originator-scoped MCP overlay (MUL-3869)

Stage 3.1 of the Composio epic (MUL-3721 parent). PR #4704 wired in the
runtime_mcp_overlay column and a per-task dispatch hook; this change
inverts the default from "all-on" to opt-in and locks the overlay to the
agent owner's own connected apps:

- Agents carry composio_toolkit_allowlist TEXT[]. NULL or [] => no MCP.
  Owner-only read/write; non-owner GET/PUT silently redacts/drops the
  field (same shape as mcp_config).
- agent_task_queue carries originator_user_id UUID. Set from the
  top-of-chain HUMAN at every enqueue path:
    * issue/mention comment by member  -> author_id
    * issue/mention comment by agent   -> inherit via comment.source_task_id
                                          -> parent task originator_user_id
    * quick-create                     -> requester_id
    * chat                             -> initiator_user_id
    * retry                            -> SQL-inherited from parent row
    * autopilot                        -> NULL (system-driven)
- BuildTaskOverlay (composio dispatch) now takes (ctx, originatorUserID,
  agent) and short-circuits on five gates: invalid originator,
  originator != agent.owner_id, empty allowlist, empty intersection of
  allowlist ∩ active connections, defensive empty session URL. Composio
  CreateSession is called with BOTH `toolkits.slugs` (the intersection)
  AND `connected_accounts` (the pinned account ids), narrowing the
  tool-router twice.
- The originator-vs-owner gate closes the agent-fanout privacy hole: any
  workspace member who can @-mention a public agent used to project the
  owner's connected apps into their run. Now the overlay only mounts
  when the human at the top of the chain IS the agent owner.

Tests:
- dispatch_test.go covers all 5 gates plus uppercase/whitespace slug
  normalisation.
- task_runtime_mcp_overlay_test.go covers the no-op gates of the new
  applyRuntimeMCPOverlay signature.
- agent_composio_allowlist_test.go (handler): owner roundtrip
  (list/empty/null), workspace-admin silent-drop, owner-only GET
  visibility, pure normaliseComposioToolkitAllowlist.
- resolve_originator_test.go (service, DB-backed): member-authored,
  agent-authored inherits via comment.source_task_id, invalid id.

Migration 129 up/down/up verified against docker postgres.

Co-authored-by: multica-agent <github@multica.ai>

* chore(composio): gofmt + regenerate sqlc with v1.31.1 (MUL-3869 review nits)

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: Eve <eve@multica-ai.local>
Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): accept nested connected account auth config

* feat(views): creator-only MCP tab for per-agent Composio allowlist (MUL-3870) (#4743)

Stage 3.2 frontend on top of the Stage 3.1 backend (MUL-3869, 4708dba97).
Adds an agent-detail tab that lets the agent owner pick which of their own
active Composio connections this agent may mount as MCP servers, writing the
selection to agent.composio_toolkit_allowlist via the existing PUT /api/agents.

- core/types: composio_toolkit_allowlist (+ _redacted) on Agent; tri-state
  composio_toolkit_allowlist on UpdateAgentRequest (omit/no-change, null/clear,
  array/replace), matching the backend contract.
- core/agents: useUpdateAgentAllowlist - optimistic mutation hook (patches the
  cached workspace agent list, rolls back on error, invalidates on settle).
- views: AgentMcpTab renders the owner's active connections as checkboxes;
  empty state links to Settings -> Integrations; defensive redacted state.
- views: wired into AgentOverviewPane as tab "composio_mcp", labeled "MCP Apps"
  to disambiguate from the existing raw-JSON "MCP" (mcp_config) tab. The entry
  is gated to the creator (currentUserId === agent.owner_id), matching the
  backend's owner-only read/write of the allowlist.
- i18n: tabs.composio_mcp + tab_body.composio_mcp.* in en/ja/ko/zh-Hans.
- tests: agent-mcp-tab.test.tsx (gating, toggle->allowlist body, active-only,
  empty, redacted); e2e/agent-mcp.spec.ts (creator sees tab + PUT body,
  non-creator hidden) with Composio + agent endpoints mocked at the boundary.

Note: the product spec says "creator"; the schema has no creator_id - the
backend gate and redaction are keyed on owner_id, so the tab uses owner_id.

Co-authored-by: multica-agent <github@multica.ai>

* fix(composio): mount remote MCP for codex

* feat(agents): agent invocation permission system (MUL-3963) (#4844)

* feat(agents): agent invocation permission system (permission_mode + invocation targets)

MUL-3963: split who may INVOKE an agent out of the overloaded visibility
column into an explicit, extensible model on feature/composio-integration.

- DB: agent.permission_mode (private|public_to) + agent_invocation_target
  table (workspace/member/team targets) + lossless backfill from visibility
  (migration 130).
- canInvokeAgent: owner-only for private (NO admin bypass, NO A2A bypass);
  public_to honours the allow-list; A2A judged by the top-of-chain originator.
- All trigger paths rewired: issue assign, comment @agent/@squad, chat,
  quick-create, autopilot, squad leader, child-done.
- Agent API: permission_mode + invocation_targets on responses and
  create/update (owner-only writes); legacy visibility kept as a derived field
  so old clients never see a permission widening.
- Composio: BuildTaskOverlay now FOLLOWS invocation permission and uses the
  agent OWNER connection (removed the originator==owner gate); front-end warns
  when a shared agent enables Composio apps.
- CLI: --permission-mode / --public-to-workspace / --public-to-member (legacy
  --visibility still mapped).
- Frontend: AccessPicker (Private / workspace / specific people / team soon),
  permission rules mirror canInvokeAgent, Composio warning banner.
- Tests: migration backfill, admin cannot invoke others private, public_to
  workspace/member whitelist, A2A by originator, Composio overlay uses owner
  connection.

Co-authored-by: multica-agent <github@multica.ai>

* feat(agents): stackable, mixed public_to invocation targets (MUL-3963)

Follow-up on PR #4844: public_to now supports selecting MULTIPLE, MIXED
targets on one agent (e.g. Public to workspace + specific people + team),
with canInvokeAgent admitting on ANY matching target (OR).

- Frontend AccessPicker: reworked from a single exclusive kind into a
  stackable multi-select — an "Everyone in workspace" toggle, a member
  multi-select checklist, and a (disabled, v1) team placeholder can be
  combined freely. Emits the full union of selected targets; empty union
  collapses to Private. Existing team targets are preserved across saves.
  Added the access.public_group locale string (en/zh-Hans/ja/ko).
- Backend already supported this (agent_invocation_target is multi-row per
  agent; create/update take a target ARRAY and batch-replace the whole
  allow-list; canInvokeAgent OR-matches). Added tests to lock it in:
  mixed member+team targets, overlapping-member batch replace, and
  workspace+member stacking then narrowing.

Refs MUL-3963.

Co-authored-by: multica-agent <github@multica.ai>

* fix(agents): address review on invocation permission (MUL-3963)

张大彪 review on PR #4844 — three blockers + product ruling + nits:

1. Migration 130: drop the FK/cascade on agent_invocation_target
   (agent_id, created_by) per the Multica no-FK rule; relationships are now
   maintained in the app layer (matching MUL-3515 §4). Added
   DeleteAgentInvocationTargetsByArchivedRuntimeAgents and call it before
   DeleteArchivedAgentsByRuntime in all three runtime-delete paths
   (runtime.go x2, runtime_profile.go) so hard-deleting agents can't orphan
   target rows.
2. revokeAndRemoveMember: prune the leaving member's member-target grants
   (DeleteAgentInvocationTargetsByMember) in the same tx as the member-row
   delete, so a re-invited user can't reclaim a stale invocation grant.
3. Empty public_to is a phantom — parsePermissionInput now normalises a
   public_to with no resolvable targets to a single workspace target, so
   `--permission-mode public_to` alone (and any empty target array) means
   "public to workspace" instead of "shared but nobody can run it".

Product ruling: the system/no-human-originator → workspace-target path in
canInvokeAgent is a deliberate, documented exception (webhook/system/
workspace-wide automation); member/team targets still fail closed without a
resolved originator. Documented in code + locked with a test.

Nits: refreshed the stale "originator must be owner" comments — models.go
(via migration 130 COMMENT ON COLUMN + sqlc regen for composio_toolkit_allowlist
and originator_user_id) and agent-mcp-tab.tsx — to the owner-connection +
invocation-permission rules.

Tests: member remove/re-add regression, system workspace exception + member
fail-closed, empty public_to → workspace (plus the earlier mixed/overlap/
batch-replace suite). Migration 130 applied to the test DB; Go handler/service/
composio suites green; views typecheck clean.

Refs MUL-3963.

Co-authored-by: multica-agent <github@multica.ai>

* fix(agents): scope member invocation-target cleanup to one workspace (MUL-3963)

张大彪 3rd review — cross-workspace permission bug + comment nits:

- DeleteAgentInvocationTargetsByMember was a GLOBAL delete by user id, so
  removing a user from workspace A also wiped their member-target grants on
  agents in workspace B. Scoped it to a single workspace by joining through
  agent.workspace_id; revokeAndRemoveMember now passes (workspaceID, userID).
- Regression test TestRevokeMember_InvocationTargetCleanupIsWorkspaceScoped:
  same user allow-listed by agents in two workspaces; removal from one leaves
  the other workspace's target intact.
- Nits: refreshed the remaining stale "originator == agent.owner_id" /
  "owner-vs-originator" comments — CreateRetryTask (agent.sql, regenerated),
  and the AgentResponse allowlist doc + ListAgents/UpdateAgent redaction
  rationale in agent.go — to the owner-connection + invocation-permission rule.

Migration 130 applied to the test DB; Go handler/service/composio suites green;
go vet clean.

Refs MUL-3963.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>

* fix(agents): agent access owner-only editable, read-only for others (MUL-3963) (#4853)

* fix(agents): make agent access owner-only editable, read-only for others (MUL-3963)

Interaction bug: a non-owner (incl. workspace admin) could open the AccessPicker
and set an agent public — the backend silently ignored it and the UI bounced
back to private. Access is owner-only, so non-owners must see a read-only state
and the backend must reject real changes explicitly.

Frontend:
- AccessPicker renders a static, non-interactive read-only state when the
  viewer is not the owner: the current access value + a lock affordance + a
  tooltip "Only the agent owner can change who can run this agent." No clickable
  trigger is rendered, so a non-owner can never open a control the backend would
  reject (the GitHub/Notion pattern for permission settings you can see but not
  edit). The editable multi-select picker is unchanged for the owner.
- agent-detail-inspector gates the picker on ownership specifically
  (currentUserId === agent.owner_id), NOT the general canEdit (which also admits
  admins, who may edit other fields but not access).
- New locale key access.owner_only_readonly (en/zh-Hans/ja/ko).

Backend:
- UpdateAgent now returns an explicit 403 when a non-owner submits a REAL
  permission change (permissionInputChangesAgent compares requested mode +
  target set against the persisted state); a no-op resubmit (admin PATCH-as-PUT
  echoing unchanged permission) is still tolerated so admin edits of other
  fields keep working. Replaces the previous silent-drop that caused the bounce.

Tests:
- access-picker.test.tsx: non-owner gets a non-interactive read-only display
  with the owner-only tooltip; owner gets an interactive picker; owner can pick
  a member and stack workspace + member.
- TestUpdateAgent_AccessChangeIsOwnerOnly: admin real change → 403; admin no-op
  resubmit → 200; admin editing other fields → 200; owner change → 200.

Incidental: fixed a pre-existing base typecheck break in
slash-command-suggestion.test.tsx (stray `signal` arg not in the suggestion
items type) that otherwise fails the whole @multica/views typecheck.

Refs MUL-3963.

Co-authored-by: multica-agent <github@multica.ai>

* fix(agents): compare legacy visibility, not expanded permission, for no-op detection (MUL-3963)

PR #4853 review: permissionInputChangesAgent expanded a legacy-only
visibility:"private" into a real private permission and compared it against the
agent's actual permission. A member-only public_to agent derives legacy
visibility "private", so an admin PATCH-as-PUT echoing visibility:"private"
while editing another field was misread as a public_to→private downgrade and
rejected with 403 — contradicting the "unchanged permission no-op is allowed"
contract.

Fix (per review): when a request carries ONLY legacy `visibility` (no
permission_mode / invocation_targets), derive the agent's CURRENT legacy
visibility from its real targets and compare the legacy string values. Equal =
no-op (allowed); a real legacy change (e.g. "workspace") still returns 403.
Requests that carry permission_mode / invocation_targets keep the precise
mode+target comparison.

Regression test TestUpdateAgent_LegacyVisibilityNoOpForMemberOnlyPublicTo:
member-only public_to agent — admin submitting visibility:"private" + a
non-permission field → 200 with targets unchanged; admin submitting
visibility:"workspace" → 403.

Go handler/composio suites green; migration 130 applied; go vet clean.

Refs MUL-3963.

Co-authored-by: multica-agent <github@multica.ai>

---------

Co-authored-by: multica-agent <github@multica.ai>

* feat(composio): brief agents on connected apps

* feat(composio): gate MCP apps behind feature flag

* fix(mobile): parse agent invocation permissions

* fix(tests): update agent fixtures for access fields

---------

Co-authored-by: multica-agent <github@multica.ai>
Co-authored-by: Multica Eve <eve@devv.ai>
Co-authored-by: Eve <eve@multica.ai>
Co-authored-by: Eve <eve@multica-ai.local>
2026-07-03 14:18:43 +08:00

1028 lines
36 KiB
Go

package handler
import (
"encoding/json"
"errors"
"log/slog"
"net/http"
"strconv"
"strings"
"time"
"github.com/go-chi/chi/v5"
"github.com/jackc/pgx/v5"
"github.com/jackc/pgx/v5/pgtype"
"github.com/multica-ai/multica/server/internal/analytics"
obsmetrics "github.com/multica-ai/multica/server/internal/metrics"
"github.com/multica-ai/multica/server/internal/middleware"
"github.com/multica-ai/multica/server/internal/util"
db "github.com/multica-ai/multica/server/pkg/db/generated"
"github.com/multica-ai/multica/server/pkg/protocol"
)
// chatSessionTitleMaxLen caps the rename input. Long enough to fit a
// meaningful summary, short enough to keep the dropdown row scannable.
const chatSessionTitleMaxLen = 200
// ---------------------------------------------------------------------------
// Chat Sessions
// ---------------------------------------------------------------------------
type CreateChatSessionRequest struct {
AgentID string `json:"agent_id"`
Title string `json:"title"`
}
func (h *Handler) CreateChatSession(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
var req CreateChatSessionRequest
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
writeError(w, http.StatusBadRequest, "invalid request body")
return
}
if req.AgentID == "" {
writeError(w, http.StatusBadRequest, "agent_id is required")
return
}
agentID, ok := parseUUIDOrBadRequest(w, req.AgentID, "agent_id")
if !ok {
return
}
workspaceUUID, ok := parseUUIDOrBadRequest(w, workspaceID, "workspace id")
if !ok {
return
}
// Verify agent exists in workspace.
agent, err := h.Queries.GetAgentInWorkspace(r.Context(), db.GetAgentInWorkspaceParams{
ID: agentID,
WorkspaceID: workspaceUUID,
})
if err != nil {
writeError(w, http.StatusNotFound, "agent not found")
return
}
if agent.ArchivedAt.Valid {
writeError(w, http.StatusBadRequest, "agent is archived")
return
}
// Invocation gate: starting a chat produces agent runs, so it uses the
// invoke permission (MUL-3963), not the softer view gate. Agent-to-agent
// chat sessions are judged by the top-of-chain originator.
actorType, actorID := h.resolveActor(r, userID, workspaceID)
if !h.canInvokeAgent(r.Context(), agent, actorType, actorID, h.invokeOriginatorFromRequest(r, actorType, actorID), workspaceID) {
writeError(w, http.StatusForbidden, "you do not have access to this agent")
return
}
session, err := h.Queries.CreateChatSession(r.Context(), db.CreateChatSessionParams{
WorkspaceID: workspaceUUID,
AgentID: agentID,
CreatorID: parseUUID(userID),
Title: req.Title,
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to create chat session")
return
}
writeJSON(w, http.StatusCreated, chatSessionToResponse(session))
}
func (h *Handler) ListChatSessions(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
// Compute the accessible-agents set once and use it to drop sessions
// whose target agent the caller no longer has access to — without this,
// a member whose role was downgraded would still see the session list
// (and transcripts via ListChatMessages) for any private agent they
// previously had access to. Falls back to the user's role from the
// workspace member context.
member, ok := h.workspaceMember(w, r, workspaceID)
if !ok {
return
}
actorType, actorID := h.resolveActor(r, userID, workspaceID)
allowed, ok := h.accessibleAgentIDs(r.Context(), workspaceID, actorType, actorID, member.Role)
if !ok {
writeError(w, http.StatusInternalServerError, "failed to resolve agent access")
return
}
status := r.URL.Query().Get("status")
// Two call sites → two row types with identical shape. Collect into a
// common response slice via small per-branch loops.
var resp []ChatSessionResponse
if status == "all" {
rows, err := h.Queries.ListAllChatSessionsByCreator(r.Context(), db.ListAllChatSessionsByCreatorParams{
WorkspaceID: parseUUID(workspaceID),
CreatorID: parseUUID(userID),
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to list chat sessions")
return
}
resp = make([]ChatSessionResponse, 0, len(rows))
for _, s := range rows {
if _, ok := allowed[uuidToString(s.AgentID)]; !ok {
continue
}
resp = append(resp, ChatSessionResponse{
ID: uuidToString(s.ID),
WorkspaceID: uuidToString(s.WorkspaceID),
AgentID: uuidToString(s.AgentID),
CreatorID: uuidToString(s.CreatorID),
Title: s.Title,
Status: s.Status,
HasUnread: s.HasUnread,
CreatedAt: timestampToString(s.CreatedAt),
UpdatedAt: timestampToString(s.UpdatedAt),
})
}
} else {
rows, err := h.Queries.ListChatSessionsByCreator(r.Context(), db.ListChatSessionsByCreatorParams{
WorkspaceID: parseUUID(workspaceID),
CreatorID: parseUUID(userID),
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to list chat sessions")
return
}
resp = make([]ChatSessionResponse, 0, len(rows))
for _, s := range rows {
if _, ok := allowed[uuidToString(s.AgentID)]; !ok {
continue
}
resp = append(resp, ChatSessionResponse{
ID: uuidToString(s.ID),
WorkspaceID: uuidToString(s.WorkspaceID),
AgentID: uuidToString(s.AgentID),
CreatorID: uuidToString(s.CreatorID),
Title: s.Title,
Status: s.Status,
HasUnread: s.HasUnread,
CreatedAt: timestampToString(s.CreatedAt),
UpdatedAt: timestampToString(s.UpdatedAt),
})
}
}
writeJSON(w, http.StatusOK, resp)
}
func (h *Handler) loadChatSessionForUser(w http.ResponseWriter, r *http.Request, userID, workspaceID, sessionID string) (db.ChatSession, bool) {
sessionUUID, ok := parseUUIDOrBadRequest(w, sessionID, "chat session id")
if !ok {
return db.ChatSession{}, false
}
workspaceUUID, ok := parseUUIDOrBadRequest(w, workspaceID, "workspace id")
if !ok {
return db.ChatSession{}, false
}
session, err := h.Queries.GetChatSessionInWorkspace(r.Context(), db.GetChatSessionInWorkspaceParams{
ID: sessionUUID,
WorkspaceID: workspaceUUID,
})
if err != nil {
writeError(w, http.StatusNotFound, "chat session not found")
return db.ChatSession{}, false
}
if uuidToString(session.CreatorID) != userID {
writeError(w, http.StatusForbidden, "not your chat session")
return db.ChatSession{}, false
}
return session, true
}
// gateChatSessionForUser combines the session ownership check with the
// private-agent access gate so a member who has lost access to the target
// agent (role downgrade, ownership transfer, agent flipped to private)
// cannot continue reading the chat transcript even though they remain the
// session creator. Returns ok=false after writing the error response.
func (h *Handler) gateChatSessionForUser(w http.ResponseWriter, r *http.Request, userID, workspaceID, sessionID string) (db.ChatSession, bool) {
session, ok := h.loadChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return db.ChatSession{}, false
}
agent, err := h.Queries.GetAgent(r.Context(), session.AgentID)
if err != nil {
writeError(w, http.StatusNotFound, "agent not found")
return db.ChatSession{}, false
}
actorType, actorID := h.resolveActor(r, userID, workspaceID)
if !h.canAccessPrivateAgent(r.Context(), agent, actorType, actorID, workspaceID) {
writeError(w, http.StatusForbidden, "you do not have access to this agent")
return db.ChatSession{}, false
}
return session, true
}
func (h *Handler) GetChatSession(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
writeJSON(w, http.StatusOK, chatSessionToResponse(session))
}
type UpdateChatSessionRequest struct {
Title *string `json:"title"`
}
// UpdateChatSession updates user-editable fields on a chat session — today
// just `title`, surfaced by the inline rename affordance in the session
// dropdown. Title is the only field accepted: `status` is legacy + read-only,
// agent/creator/workspace are immutable, the resume pointers
// (session_id / work_dir / runtime_id) are daemon-owned.
func (h *Handler) UpdateChatSession(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
var req UpdateChatSessionRequest
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
writeError(w, http.StatusBadRequest, "invalid request body")
return
}
if req.Title == nil {
writeError(w, http.StatusBadRequest, "title is required")
return
}
title := strings.TrimSpace(*req.Title)
if title == "" {
writeError(w, http.StatusBadRequest, "title is required")
return
}
if len([]rune(title)) > chatSessionTitleMaxLen {
writeError(w, http.StatusBadRequest, "title is too long")
return
}
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
updated, err := h.Queries.UpdateChatSessionTitle(r.Context(), db.UpdateChatSessionTitleParams{
ID: session.ID,
Title: title,
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to update chat session")
return
}
resolvedSessionID := uuidToString(updated.ID)
h.publishChat(protocol.EventChatSessionUpdated, workspaceID, "member", userID, resolvedSessionID, protocol.ChatSessionUpdatedPayload{
ChatSessionID: resolvedSessionID,
Title: updated.Title,
UpdatedAt: timestampToString(updated.UpdatedAt),
})
writeJSON(w, http.StatusOK, chatSessionToResponse(updated))
}
// DeleteChatSession hard-deletes a chat session owned by the caller. The
// row lock + cancel + delete run inside a single tx so a concurrent
// SendChatMessage cannot enqueue a task that would later be orphaned by
// the FK ON DELETE SET NULL on agent_task_queue.chat_session_id. Cancel
// failure aborts the delete; events fire only after commit.
func (h *Handler) DeleteChatSession(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.loadChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
tx, err := h.TxStarter.Begin(r.Context())
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to start transaction")
return
}
defer tx.Rollback(r.Context())
qtx := h.Queries.WithTx(tx)
// FOR UPDATE on the chat_session row blocks any concurrent INSERT into
// agent_task_queue that references it (the FK validation needs a
// KEY SHARE lock). After we commit the delete, the blocked INSERT
// fails its FK check, so it can't land an orphaned task.
if _, err := qtx.LockChatSessionForDelete(r.Context(), session.ID); err != nil {
if errors.Is(err, pgx.ErrNoRows) {
// Already gone — treat as idempotent success.
w.WriteHeader(http.StatusNoContent)
return
}
writeError(w, http.StatusInternalServerError, "failed to lock chat session")
return
}
cancelled, err := qtx.CancelAgentTasksByChatSession(r.Context(), session.ID)
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to cancel chat session tasks")
return
}
// channel_chat_session_binding used to carry a chat_session FK with
// ON DELETE CASCADE; MUL-3515 §4 dropped every channel_* foreign key, so
// prune the binding here in the same tx that deletes its chat_session.
if err := qtx.DeleteChannelChatSessionBindingBySession(r.Context(), session.ID); err != nil {
writeError(w, http.StatusInternalServerError, "failed to delete chat session binding")
return
}
if err := qtx.DeleteChatSession(r.Context(), db.DeleteChatSessionParams{
ID: session.ID,
WorkspaceID: session.WorkspaceID,
}); err != nil {
writeError(w, http.StatusInternalServerError, "failed to delete chat session")
return
}
if err := tx.Commit(r.Context()); err != nil {
slog.Warn("commit chat session delete failed", "session_id", sessionID, "error", err)
writeError(w, http.StatusInternalServerError, "failed to commit chat session delete")
return
}
// Post-commit broadcasts. Subscribers should never observe events for a
// tx that didn't actually persist.
h.TaskService.BroadcastCancelledTasks(r.Context(), cancelled)
resolvedSessionID := uuidToString(session.ID)
h.publishChat(protocol.EventChatSessionDeleted, workspaceID, "member", userID, resolvedSessionID, protocol.ChatSessionDeletedPayload{
ChatSessionID: resolvedSessionID,
})
w.WriteHeader(http.StatusNoContent)
}
// ---------------------------------------------------------------------------
// Chat Messages
// ---------------------------------------------------------------------------
type SendChatMessageRequest struct {
Content string `json:"content"`
AttachmentIDs []string `json:"attachment_ids"`
}
type SendChatMessageResponse struct {
MessageID string `json:"message_id"`
TaskID string `json:"task_id"`
// AttachmentIDs are the attachment rows actually bound to this message by
// the server. The client diffs these against the ids it requested so it
// can warn the user when an attachment silently failed to bind — no extra
// round-trip needed. No `omitempty`: a send that requested attachments but
// bound none must serialize `[]` (not be omitted), otherwise the client
// can't tell "all binds failed" from "older server without this field" and
// would silently skip the very warning this exists for. When no
// attachments were requested the value is nil → `null`, which the client's
// guard short-circuits on the requested-ids check.
AttachmentIDs []string `json:"attachment_ids"`
// CreatedAt anchors the chat StatusPill timer the instant the user
// hits send. Without it the front-end falls back to its local clock
// and the timer "snaps backwards" later when WS events deliver the
// real created_at. Returning it here means the pill renders 0s from
// the start with a stable anchor.
CreatedAt string `json:"created_at"`
}
func (h *Handler) SendChatMessage(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
var req SendChatMessageRequest
if err := json.NewDecoder(r.Body).Decode(&req); err != nil {
writeError(w, http.StatusBadRequest, "invalid request body")
return
}
if req.Content == "" {
writeError(w, http.StatusBadRequest, "content is required")
return
}
// Pre-validate attachment ids early so invalid input returns 400 before
// any state mutation. The actual link runs after CreateChatMessage so we
// have a message_id to back-fill into the attachment rows.
attachmentIDs, ok := parseUUIDSliceOrBadRequest(w, req.AttachmentIDs, "attachment_ids")
if !ok {
return
}
// Load chat session and re-check the private-agent gate on every send.
// The session's creator passed the gate at create time, but their
// workspace role (or the agent's owner) may have changed since — keep
// stale sessions from being a back-door into a private agent the user
// can no longer reach. Agent senders bypass to preserve A2A collaboration.
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
// New archive flow doesn't exist anymore, but legacy rows with
// status='archived' may still be in the DB from before the feature
// was removed. Refuse to enqueue new agent work for them — frontend
// surfaces these as read-only.
if session.Status != "active" {
writeError(w, http.StatusBadRequest, "chat session is archived")
return
}
// Create the user message first so the daemon can always find it.
msg, err := h.Queries.CreateChatMessage(r.Context(), db.CreateChatMessageParams{
ChatSessionID: session.ID,
Role: "user",
Content: req.Content,
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to create chat message")
return
}
// Back-fill chat_message_id on attachments the sender uploaded while
// composing. New clients upload workspace-scoped unattached rows and bind
// them here; older clients may still upload against the chat_session_id.
// The query accepts both shapes, but only for this workspace, this actor,
// and rows that are not already linked to an issue/comment/message.
var boundAttachmentIDs []string
if len(attachmentIDs) > 0 {
actorType, actorID := h.resolveActor(r, userID, workspaceID)
bound, err := h.Queries.LinkAttachmentsToChatMessage(r.Context(), db.LinkAttachmentsToChatMessageParams{
ChatMessageID: msg.ID,
ChatSessionID: session.ID,
WorkspaceID: session.WorkspaceID,
UploaderType: actorType,
UploaderID: parseUUID(actorID),
AttachmentIds: attachmentIDs,
})
if err != nil {
// Don't fail the send — the message content is already saved and
// the attachments remain on the session (still downloadable).
slog.Warn("link chat attachments failed", "error", err, "message_id", uuidToString(msg.ID))
}
boundAttachmentIDs = make([]string, 0, len(bound))
for _, id := range bound {
boundAttachmentIDs = append(boundAttachmentIDs, uuidToString(id))
}
}
// Enqueue a chat task after the message exists. For web chat the sender is
// the authenticated request user (sessions are creator-only), so they are
// the task initiator — surfaced to the agent under `## Task Initiator`.
task, err := h.TaskService.EnqueueChatTask(r.Context(), session, parseUUID(userID), false)
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to enqueue chat task: "+err.Error())
return
}
if err := h.Queries.LinkChatMessageToTask(r.Context(), db.LinkChatMessageToTaskParams{
ID: msg.ID,
TaskID: task.ID,
}); err != nil {
// Don't fail the send: the task already exists and the user message
// is persisted. The link is only needed for precise empty-cancel
// cleanup; older/unlinked rows simply keep the historical behavior.
slog.Warn("link user chat message to task failed",
"message_id", uuidToString(msg.ID),
"task_id", uuidToString(task.ID),
"error", err,
)
}
// Touch session updated_at.
if err := h.Queries.TouchChatSession(r.Context(), session.ID); err != nil {
slog.Warn("failed to touch chat session", "session_id", sessionID, "error", err)
}
taskContext := h.TaskService.AnalyticsContextForTask(r.Context(), task)
platform, _, _ := middleware.ClientMetadataFromContext(r.Context())
obsmetrics.RecordEvent(h.Analytics, h.Metrics, analytics.ChatMessageSent(
userID,
workspaceID,
uuidToString(session.ID),
uuidToString(task.ID),
uuidToString(session.AgentID),
taskContext.RuntimeMode,
taskContext.Provider,
platform,
))
// Broadcast the user message.
resolvedSessionID := uuidToString(session.ID)
h.publishChat(protocol.EventChatMessage, workspaceID, "member", userID, resolvedSessionID, protocol.ChatMessagePayload{
ChatSessionID: resolvedSessionID,
MessageID: uuidToString(msg.ID),
Role: "user",
Content: req.Content,
TaskID: uuidToString(task.ID),
CreatedAt: timestampToString(msg.CreatedAt),
})
writeJSON(w, http.StatusCreated, SendChatMessageResponse{
MessageID: uuidToString(msg.ID),
TaskID: uuidToString(task.ID),
CreatedAt: timestampToString(task.CreatedAt),
AttachmentIDs: boundAttachmentIDs,
})
}
type ChatMessagesCursorResponse struct {
CreatedAt string `json:"created_at"`
ID string `json:"id"`
}
type ChatMessagesPageResponse struct {
Messages []ChatMessageResponse `json:"messages"`
Limit int `json:"limit"`
HasMore bool `json:"has_more"`
NextCursor *ChatMessagesCursorResponse `json:"next_cursor,omitempty"`
}
func parseChatMessagesPageParams(r *http.Request) (int, pgtype.Timestamptz, pgtype.UUID, error) {
limit := 50
if raw := r.URL.Query().Get("limit"); raw != "" {
parsed, err := strconv.Atoi(raw)
if err != nil || parsed < 1 || parsed > 100 {
return 0, pgtype.Timestamptz{}, pgtype.UUID{}, errors.New("invalid limit")
}
limit = parsed
}
rawBeforeCreatedAt := r.URL.Query().Get("before_created_at")
rawBeforeID := r.URL.Query().Get("before_id")
if rawBeforeCreatedAt == "" && rawBeforeID == "" {
return limit, pgtype.Timestamptz{}, pgtype.UUID{}, nil
}
if rawBeforeCreatedAt == "" || rawBeforeID == "" {
return 0, pgtype.Timestamptz{}, pgtype.UUID{}, errors.New("invalid cursor")
}
beforeTime, err := time.Parse(time.RFC3339Nano, rawBeforeCreatedAt)
if err != nil {
return 0, pgtype.Timestamptz{}, pgtype.UUID{}, errors.New("invalid cursor")
}
beforeID, err := util.ParseUUID(rawBeforeID)
if err != nil {
return 0, pgtype.Timestamptz{}, pgtype.UUID{}, errors.New("invalid cursor")
}
return limit, pgtype.Timestamptz{Time: beforeTime, Valid: true}, beforeID, nil
}
func (h *Handler) ListChatMessages(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
messages, err := h.Queries.ListChatMessages(r.Context(), session.ID)
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to list chat messages")
return
}
messageIDs := make([]pgtype.UUID, len(messages))
for i, m := range messages {
messageIDs[i] = m.ID
}
groupedAtt := h.groupChatMessageAttachments(r.Context(), workspaceID, messageIDs)
resp := make([]ChatMessageResponse, len(messages))
for i, m := range messages {
resp[i] = chatMessageToResponse(m, groupedAtt[uuidToString(m.ID)])
}
writeJSON(w, http.StatusOK, resp)
}
func (h *Handler) ListChatMessagesPage(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
limit, beforeCreatedAt, beforeID, err := parseChatMessagesPageParams(r)
if err != nil {
writeError(w, http.StatusBadRequest, err.Error())
return
}
messages, err := h.Queries.ListChatMessagesPage(r.Context(), db.ListChatMessagesPageParams{
ChatSessionID: session.ID,
Limit: int32(limit + 1),
BeforeCreatedAt: beforeCreatedAt,
BeforeID: beforeID,
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to list chat messages")
return
}
hasMore := len(messages) > limit
if hasMore {
messages = messages[:limit]
}
var nextCursor *ChatMessagesCursorResponse
if hasMore && len(messages) > 0 {
oldest := messages[len(messages)-1]
nextCursor = &ChatMessagesCursorResponse{
CreatedAt: oldest.CreatedAt.Time.Format(time.RFC3339Nano),
ID: uuidToString(oldest.ID),
}
}
// SQL fetches newest windows first so the empty cursor opens at the recent
// tail. Reverse each cursor page before serializing to keep message order
// chronological within the viewport.
for i, j := 0, len(messages)-1; i < j; i, j = i+1, j-1 {
messages[i], messages[j] = messages[j], messages[i]
}
messageIDs := make([]pgtype.UUID, len(messages))
for i, m := range messages {
messageIDs[i] = m.ID
}
groupedAtt := h.groupChatMessageAttachments(r.Context(), workspaceID, messageIDs)
resp := make([]ChatMessageResponse, len(messages))
for i, m := range messages {
resp[i] = chatMessageToResponse(m, groupedAtt[uuidToString(m.ID)])
}
writeJSON(w, http.StatusOK, ChatMessagesPageResponse{
Messages: resp,
Limit: limit,
HasMore: hasMore,
NextCursor: nextCursor,
})
}
// PendingChatTaskResponse is returned by GetPendingChatTask — either the
// current in-flight task's id/status, or an empty object when none is active.
// CreatedAt is the anchor the frontend uses to time the chat StatusPill
// (elapsed seconds = now - CreatedAt). It must come from the server because
// optimistic seeds don't have a real task created_at and the timer needs to
// survive refresh / reopen.
type PendingChatTaskResponse struct {
TaskID string `json:"task_id,omitempty"`
Status string `json:"status,omitempty"`
CreatedAt string `json:"created_at,omitempty"`
}
// MarkChatSessionRead clears the session's unread_since (→ has_unread=false)
// and broadcasts chat:session_read so other devices of the same user drop
// their badges.
func (h *Handler) MarkChatSessionRead(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
if err := h.Queries.MarkChatSessionRead(r.Context(), session.ID); err != nil {
writeError(w, http.StatusInternalServerError, "failed to mark session read")
return
}
resolvedSessionID := uuidToString(session.ID)
h.publishChat(protocol.EventChatSessionRead, workspaceID, "member", userID, resolvedSessionID, protocol.ChatSessionReadPayload{
ChatSessionID: resolvedSessionID,
})
w.WriteHeader(http.StatusNoContent)
}
// PendingChatTasksResponse is the aggregate view consumed by the FAB.
type PendingChatTasksResponse struct {
Tasks []PendingChatTaskItem `json:"tasks"`
}
type PendingChatTaskItem struct {
TaskID string `json:"task_id"`
Status string `json:"status"`
ChatSessionID string `json:"chat_session_id"`
}
type CancelledChatMessageResponse struct {
ChatSessionID string `json:"chat_session_id"`
MessageID string `json:"message_id"`
Content string `json:"content"`
RestoreToInput bool `json:"restore_to_input"`
Attachments []AttachmentResponse `json:"attachments,omitempty"`
}
type CancelTaskByUserResponse struct {
AgentTaskResponse
CancelledChatMessage *CancelledChatMessageResponse `json:"cancelled_chat_message,omitempty"`
}
// ListPendingChatTasks returns every in-flight chat task owned by the current
// user in this workspace. Drives the FAB's "running" indicator when the chat
// window is closed (no per-session query is subscribed). Tasks belonging to
// private agents the caller has lost access to are dropped from the response.
func (h *Handler) ListPendingChatTasks(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
member, ok := h.workspaceMember(w, r, workspaceID)
if !ok {
return
}
actorType, actorID := h.resolveActor(r, userID, workspaceID)
allowed, ok := h.accessibleAgentIDs(r.Context(), workspaceID, actorType, actorID, member.Role)
if !ok {
writeError(w, http.StatusInternalServerError, "failed to resolve agent access")
return
}
rows, err := h.Queries.ListPendingChatTasksByCreator(r.Context(), db.ListPendingChatTasksByCreatorParams{
WorkspaceID: parseUUID(workspaceID),
CreatorID: parseUUID(userID),
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to list pending chat tasks")
return
}
// Map session → agent so we can filter without an N+1. The user's own
// session list is small, so one extra query is cheaper than per-row
// lookups.
sessions, err := h.Queries.ListAllChatSessionsByCreator(r.Context(), db.ListAllChatSessionsByCreatorParams{
WorkspaceID: parseUUID(workspaceID),
CreatorID: parseUUID(userID),
})
if err != nil {
writeError(w, http.StatusInternalServerError, "failed to resolve chat session agents")
return
}
sessionAgent := make(map[string]string, len(sessions))
for _, s := range sessions {
sessionAgent[uuidToString(s.ID)] = uuidToString(s.AgentID)
}
items := make([]PendingChatTaskItem, 0, len(rows))
for _, row := range rows {
sessionID := uuidToString(row.ChatSessionID)
agentID, hasAgent := sessionAgent[sessionID]
if !hasAgent {
continue
}
if _, ok := allowed[agentID]; !ok {
continue
}
items = append(items, PendingChatTaskItem{
TaskID: uuidToString(row.TaskID),
Status: row.Status,
ChatSessionID: sessionID,
})
}
writeJSON(w, http.StatusOK, PendingChatTasksResponse{Tasks: items})
}
// GetPendingChatTask returns the most recent in-flight task (queued / dispatched
// / running) for a chat session. The frontend polls this on mount / session
// switch so pending UI state survives refresh and reopen.
func (h *Handler) GetPendingChatTask(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
sessionID := chi.URLParam(r, "sessionId")
session, ok := h.gateChatSessionForUser(w, r, userID, workspaceID, sessionID)
if !ok {
return
}
task, err := h.Queries.GetPendingChatTask(r.Context(), session.ID)
if err != nil {
// No in-flight task — return an empty object, not an error.
writeJSON(w, http.StatusOK, PendingChatTaskResponse{})
return
}
writeJSON(w, http.StatusOK, PendingChatTaskResponse{
TaskID: uuidToString(task.ID),
Status: task.Status,
CreatedAt: timestampToString(task.CreatedAt),
})
}
// ---------------------------------------------------------------------------
// Task cancellation (user-facing, with ownership check)
// ---------------------------------------------------------------------------
// CancelTaskByUser cancels a task the caller is allowed to act on within the
// current workspace.
//
// Tenancy is enforced uniformly through the task's owning agent: every
// agent_task_queue row carries a NOT NULL agent_id (ON DELETE CASCADE, so the
// agent always exists), and agents are workspace-scoped. GetAgentTaskInWorkspace
// is therefore the single tenant guard that works regardless of which optional
// source FK (issue / chat_session / autopilot_run) is set — which is what makes
// run_only autopilot tasks and quick_create tasks (whose issue does not exist
// yet) cancellable at all. Keying cancellation off issue_id / chat_session_id
// alone is exactly what 404'd these tasks before (MUL-2827).
//
// On top of tenancy, two privacy models layer on:
// - a chat task is private to the member who started the conversation, so
// only that creator may cancel it;
// - every other task surfaces on the agent Activity tab and the workspace
// task snapshot, both of which hide private agents from members without
// access. Cancellation mirrors that gate via canAccessPrivateAgent so the
// id-only endpoint is never more permissive than the surface that exposes
// the task.
func (h *Handler) CancelTaskByUser(w http.ResponseWriter, r *http.Request) {
userID, ok := requireUserID(w, r)
if !ok {
return
}
workspaceID := ctxWorkspaceID(r.Context())
wsUUID, ok := parseUUIDOrBadRequest(w, workspaceID, "workspace id")
if !ok {
return
}
taskID := chi.URLParam(r, "taskId")
taskUUID, ok := parseUUIDOrBadRequest(w, taskID, "task id")
if !ok {
return
}
task, err := h.Queries.GetAgentTaskInWorkspace(r.Context(), db.GetAgentTaskInWorkspaceParams{
ID: taskUUID,
WorkspaceID: wsUUID,
})
if err != nil {
writeError(w, http.StatusNotFound, "task not found")
return
}
if task.ChatSessionID.Valid {
// Chat privacy: only the member who opened the conversation may
// cancel its task, even though the workspace is shared.
cs, err := h.Queries.GetChatSessionInWorkspace(r.Context(), db.GetChatSessionInWorkspaceParams{
ID: task.ChatSessionID,
WorkspaceID: wsUUID,
})
if err != nil {
writeError(w, http.StatusNotFound, "task not found")
return
}
if uuidToString(cs.CreatorID) != userID {
writeError(w, http.StatusForbidden, "not your task")
return
}
} else {
// Issue / autopilot / quick_create tasks are all visible on the
// agent Activity tab + workspace snapshot, which gate private
// agents. Mirror that gate here.
agent, err := h.Queries.GetAgentInWorkspace(r.Context(), db.GetAgentInWorkspaceParams{
ID: task.AgentID,
WorkspaceID: wsUUID,
})
if err != nil {
writeError(w, http.StatusNotFound, "task not found")
return
}
actorType, actorID := h.resolveActor(r, userID, workspaceID)
if !h.canAccessPrivateAgent(r.Context(), agent, actorType, actorID, workspaceID) {
writeError(w, http.StatusForbidden, "you do not have access to this agent")
return
}
}
cancelled, err := h.TaskService.CancelTaskWithResult(r.Context(), taskUUID)
if err != nil {
writeError(w, http.StatusBadRequest, err.Error())
return
}
resp := CancelTaskByUserResponse{
AgentTaskResponse: taskToResponse(cancelled.Task, workspaceID),
}
if cancelled.CancelledChatMessage != nil {
attachments := make([]AttachmentResponse, 0, len(cancelled.CancelledChatMessage.Attachments))
for _, a := range cancelled.CancelledChatMessage.Attachments {
attachments = append(attachments, h.attachmentToResponse(a))
}
resp.CancelledChatMessage = &CancelledChatMessageResponse{
ChatSessionID: cancelled.CancelledChatMessage.ChatSessionID,
MessageID: cancelled.CancelledChatMessage.MessageID,
Content: cancelled.CancelledChatMessage.Content,
RestoreToInput: cancelled.CancelledChatMessage.RestoreToInput,
Attachments: attachments,
}
}
writeJSON(w, http.StatusOK, resp)
}
// ---------------------------------------------------------------------------
// Response types & helpers
// ---------------------------------------------------------------------------
type ChatSessionResponse struct {
ID string `json:"id"`
WorkspaceID string `json:"workspace_id"`
AgentID string `json:"agent_id"`
CreatorID string `json:"creator_id"`
Title string `json:"title"`
Status string `json:"status"`
// Only populated by list endpoints — single-session fetches return false.
HasUnread bool `json:"has_unread"`
CreatedAt string `json:"created_at"`
UpdatedAt string `json:"updated_at"`
}
type ChatMessageResponse struct {
ID string `json:"id"`
ChatSessionID string `json:"chat_session_id"`
Role string `json:"role"`
Content string `json:"content"`
TaskID *string `json:"task_id"`
CreatedAt string `json:"created_at"`
// FailureReason flags an assistant row synthesized by FailTask's chat
// fallback. Front-end uses it to switch to the destructive bubble.
FailureReason *string `json:"failure_reason"`
// ElapsedMs is the wall-clock duration from task creation to terminal
// state. Drives "Replied in 38s" / "Failed after 12s" captions.
ElapsedMs *int64 `json:"elapsed_ms"`
// Attachments linked to this message via chat_message_id. The chat
// bubble renders file cards from these, and the daemon claim path
// (daemon.go) pulls structured metadata from the same source so the
// agent can `multica attachment download <id>` rather than guessing
// from a markdown URL that may expire.
Attachments []AttachmentResponse `json:"attachments,omitempty"`
}
func chatSessionToResponse(s db.ChatSession) ChatSessionResponse {
return ChatSessionResponse{
ID: uuidToString(s.ID),
WorkspaceID: uuidToString(s.WorkspaceID),
AgentID: uuidToString(s.AgentID),
CreatorID: uuidToString(s.CreatorID),
Title: s.Title,
Status: s.Status,
CreatedAt: timestampToString(s.CreatedAt),
UpdatedAt: timestampToString(s.UpdatedAt),
}
}
func chatMessageToResponse(m db.ChatMessage, attachments []AttachmentResponse) ChatMessageResponse {
return ChatMessageResponse{
ID: uuidToString(m.ID),
ChatSessionID: uuidToString(m.ChatSessionID),
Role: m.Role,
Content: m.Content,
TaskID: uuidToPtr(m.TaskID),
CreatedAt: timestampToString(m.CreatedAt),
FailureReason: textToPtr(m.FailureReason),
ElapsedMs: int8ToPtr(m.ElapsedMs),
Attachments: attachments,
}
}