Skip to content
← Back to changelog index

3.1.4

Nov 09, 2026 · 17 packages · 30 unique changes · 11 release entries

appsbundlesintegrationslibsmodules

This release affects the integrations, sharedLibs, solutions families.

Run contractspec connect adoption resolve --family integrations to see how it impacts your project.

Release summaries

  • cmp-execution-lanes-delegation

    Team workers run under bounded, least-privileged delegation with typed help and two-phase lifecycle work orders, and team readiness and evidence follow the finalization rules

    integrator

    createTeamRun accepts an optional delegation authority (TeamDelegationAuthorityV1 or a per-run resolver) and workOrderTimeoutMs. Every team worker is launched with a TeamWorkerDelegationV1 projection (derived principal, objective plus claimable tasks only, role tools within the lead's tools, role-required grants only, lease wall-clock bound, no further delegation) in TeamBackendWorkerLaunchV1, and every backend lifecycle call receives its work-order id and deadline. Under an authority, worker-scoped operations must pass the worker's derived principal as actorId. Evidence bundle ids must be saved to the lane store before they are attached, every new run must enable the verification lane, and a worker only claims tasks hinted for its role or unhinted. Typed help uses the canonical agent-protocol help request and response. evaluateLaneReadiness no longer reports ready on a verifier self-signoff.

    agent

    A team worker acts only as its derived principal, sees only the run objective and the tasks it may claim, holds only its role's tools within the lead's and the grants its role requires, and cannot delegate further. Ask for help with requestHelp and a canonical AgentHelpRequestV1 instead of guessing; the lead, or a responder the host help-response authority verifies, answers with respondToHelp. No worker of the run may answer.

    maintainer

    Team readiness lives in runtime/team-readiness.ts and shares its approval rules with finalization (lanes/team/finalize.ts); it runs for every snapshot with team state and resolves keyed lane policies through policyRegistry or the default team.coordinated policy. Delegation projections are derived in lanes/team/team-delegation.ts and enforced in team-delegation-authority.ts (including the run's delegation binding); typed help is in team-help.ts and team-help-response.ts; backend lifecycle calls go only through runTeamWorkerLifecycle (team-work-order-runner.ts, team-work-orders.ts, team-work-order-store.ts, team-work-order-apply.ts), never inside withExclusiveTeamMutation. The ledger row delegation.bounded-and-projected is implemented with residual blockers, and completion.evidence-backed names its remaining legacy gap.

  • cmp-fix-w8-delegation-boundary-composition

    Preserve persisted team authority and the exact bounded projection through concrete backends

    integrator

    Reopening a delegation-bound team run requires its current host authority, even when the new runtime omits the delegation source. Never-bound runs retain legacy behavior. Queue and workflow worker payloads preserve the existing delegation, workOrderId and deadlineAt without forwarding whole launch objects, descriptors, provider handles or unrelated parent context.

    maintainer

    This patch closes the missing-source and concrete-adapter gaps recorded in the earlier cmp-execution-lanes-delegation capsule. Public optional field types and lifecycle run/worker handle maps remain unchanged. No production host, grant or worker execution is activated.

  • communication-agent-protocol-owner

    Communication projections use the canonical pure help definition

    integrator

    Existing Communication lifecycle projections and validation remain compatible with canonical durable help request types.

  • communication-os-ai-gateway-models

    CommunicationOS thread summaries and AI reply drafts run only through the Vercel AI Gateway. Users can switch between allowlisted Anthropic, OpenAI and Google models per request, and governance managers set the workspace default.

    integrator

    comms.thread.summarize and comms.reply.draft 1.1.0 accept an optional allowlisted Gateway model. The change is additive, with the named errors AI_MODEL_NOT_ALLOWED and AI_MODEL_OVERRIDE_DENIED. COMMUNICATION_OS_AI_MODEL_CATALOG is the single allowlist. communication.ai.model-settings.get|set and the communication.ai.model-setting.changed event define the workspace setting. communication-os-agent adds ./thread-assist-gateway over createVercelGatewayModel. The client adds getAiModelSettings, setAiModelSettings and CommunicationOsApiError.code.

    maintainer

    communicationos-api reads COMMUNICATION_OS_AI_ASSIST_ENABLED=true plus Gateway authentication (AI_GATEWAY_API_KEY, or the per-invocation Vercel OIDC token header), read per request; /v2/health reports aiAssist. COMMUNICATION_OS_AI_DEFAULT_MODEL is optional and must be allowlisted. Direct provider keys (MISTRAL_API_KEY, ANTHROPIC_API_KEY, COMMOS_LLM_*) are ignored. The Gateway requests EU-zone inference, zero data retention and no training. Gateway request processing itself is global.

    customer

    On a CompanyOS thread, users with AI access can choose which model writes the summary or the reply draft. Each result says which model generated it. Workspace administrators choose the default model on the CommunicationOS governance page. Drafts still need a person to approve and send them.

  • communication-os-ai-inbox-triage

    CommunicationOS inbox threads can carry advisory AI triage labels (category, urgency, needs reply, reason) produced through the Vercel AI Gateway. Labels are cached per thread revision, journaled with their model and shown as marked AI chips with filters in the CompanyOS inbox.

    integrator

    New additive operations comms.thread.triage and comms.thread.triage.batch (1.0.0) plus the communication.ai.thread-triaged event. communication-os.ai.use (2.1.0) provides them. The label taxonomy, strict parser and structured-output JSON schema live in @lssm-tech/lib.communication-spec/contracts/thread-triage. communication-os-agent adds ./thread-triage and a triage() method on the thread assist port. The client adds triageThread and triageThreads. ThreadListItem.aiTriage is optional, so existing hosts are unaffected.

    maintainer

    No new environment variables. Triage uses the existing COMMUNICATION_OS_AI_ASSIST_ENABLED gate, Gateway authentication and model resolution. Labels are stored as communication.ai.thread-triaged journal events, which carry no roomId, so room and message projections are unchanged. The per-principal rate limit (100 model calls per 10 minutes) is kept in process memory, so on serverless it applies per instance.

    customer

    In the CompanyOS CommunicationOS inbox, users with AI access see AI-marked chips on each conversation (category, urgency, needs reply). Hovering shows why and which model suggested it. They can filter by AI category or by whether a reply is needed. The labels are only suggestions: nothing is sent, assigned or changed automatically.

  • communication-os-live-channels-governance

    CommunicationOS provider connections, room bindings, governance policies, contact identities and audit export inside CompanyOS

    integrator

    Additive contracts: communication.audit.export@1.0.0 (query) and communication.audit.exported@1.0.0 (event), plus governance document models for retention, consent, escalation, binding consent and identity review evidence stored in governed-work details. startProviderOAuth accepts providerKey.

    customer

    CompanyOS gains /companyos/communication-os/connections (provider OAuth return, status refresh, room binding with recorded consent), governance policy publishing, per-network contact identities with merge evidence, and a bounded CSV/JSON journal export that is journaled before download.

  • communication-os-meeting-threads

    Meetings join CommunicationOS in CompanyOS: an OPA meeting gets one CommunicationOS conversation with a shared agenda, participants, shared notes and governed follow-ups. The hCircle meeting host becomes production-qualifiable through a startup configuration verdict, a health field, a proven schema ledger and an exact-SHA canary.

    integrator

    New additive operations communication.meeting_thread.list, .link and .revise (1.0.0, experimental) with COMMUNICATION_MEETING_THREAD_LIMITS. A meeting thread links one OPA meeting (opa_smart_meeting or opa_meeting_workspace) to one room; follow-ups use the existing communication.governed_work.create with kind action, threadId set to the meeting room and details.meetingThreadId. The runtime adds ./ecosystem-meeting-thread, the client adds listMeetingThreads, linkMeetingThread and reviseMeetingThread, and the module adds the MeetingList and MeetingThreadDetail organisms. Existing contracts are unchanged.

    maintainer

    Meeting threads need no new environment variables or SQL migrations: they are journal events in the existing communication_os journal. The hCircle meeting host stays off unless HCIRCLE_MEETING_RUNTIME_ENABLED=true. When enabled it validates HCIRCLE_GOVERNED_DATA_DATABASE_URL, HCIRCLE_GOVERNED_DATA_MIGRATION_URL (a separate migrator connection), DATABASE_URL, HCIRCLE_MEETING_BYOK_MASTER_KEY and OPA_MEETING_ACCESS_GRANT_PEPPER (each at least 32 characters and different), HCIRCLE_GOVERNED_DATA_SSL_MODE and COMMUNICATION_OS_API_ORIGIN. HCIRCLE_GOVERNED_RECOVERY_TOKEN (at least 32 characters) is also required because governed-data recovery shares HCIRCLE_GOVERNED_DATA_DATABASE_URL and refuses startup without it. /v2/health reports meetingHost and release.sha; a misconfigured host answers 503 on meeting routes instead of aborting startup. The hcircle_meeting_schema_ledger still applies 0039, 0040 and 0041 and now requires the opa_app and opa_migrator roles beforehand and proves forced row-level security on all 21 runtime tables.

    customer

    In CompanyOS, CommunicationOS > Meetings lists the workspace meetings planned in OPA. Team members with reply access start the meeting conversation, keep a shared agenda, participants and notes, and create follow-ups that are reviewed before anything is sent. Preparation, private notes, transcripts and outcomes stay in the OPA meeting workspace, one click away.

  • red-gates-dependency-hygiene

    Communication spec drops its ProviderSpec dependency and two integrations publish caret workspace ranges; no API changes.

    integrator

    @lssm-tech/lib.communication-spec no longer installs @lssm-tech/lib.provider-spec. ./contracts/ai-model-catalog keeps the same exports and the same seven allowlisted Gateway models. The entries are now plain data owned by the spec instead of aliases of the provider-spec objects, so compare entries by value, not by reference. provider-linkedin and runtime-local publish caret ranges for their internal dependencies.

  • sms-consent-stop-start

    SMS STOP/START/HELP keywords become CAS preference revisions with sms_opt_out evidence, and outbound SMS fails closed without durable consent

    integrator

    Inbound STOP, START and HELP (English and French) normalize into keyword signals. STOP and START become compare-and-swap preference revisions in the scoped CommunicationOS journal. The latest revision produces the sms_opt_out evidence that delivery eligibility requires. The delivery dispatcher refuses SMS without an smsConsent gate or an opted-in state. The owner confirmed minor rather than major (release decision B021, 2026-10-26): no in-repo production path dispatches SMS through createCommunicationDeliveryDispatcher (communicationos-worker sends through the Runtime Node delivery port, and createEcosystemDeliveryDispatcher is used only by tests), and sending SMS without durable consent evidence is the compliance defect this change closes.

  • taketide-public-overview-ownership

    Keep public overview rendering in the neutral creator-platform experience and brand content in Taketide composition.

    integrator

    Use the explicit PublicOverview.web entry with a data-only brand composition; generic and native entries remain unchanged.

  • ts-01-seq

    Tellshift sequences, consent-bound email enrollment, durable due discovery, approved Gmail send and tenant erasure on CommunicationOS

    integrator

    Run the CommunicationOS migrations to reach 0016_tellshift_sequences_rls. Due discovery calls communication_os.due_sequence_tenants(now) as communication_os_app and receives tenant ids only; all per-tenant work must use the tenant-scoped provider. Wire the Gmail sequence adapter with a grant source and a recipient resolver; uncertain sends are reconciled with the adapter's read-only reconcile.

    maintainer

    A job sends only its exact approved bytes once: consent, revision approval and policy fingerprint, redaction, content proof, reply stop, reply coverage and credential are rechecked before the claim. A reply stops the sequence; holds never auto-resume; an uncertain send is never resent. Consumers that iterated COMMUNICATION_OS_RLS_TABLES now also see the three sequence tables. The managed migration owner must not be a superuser for RLS to be meaningful; tests prove the path under a NOSUPERUSER owner.

Deprecations

  • communicationos-api no longer uses MISTRAL_API_KEY, ANTHROPIC_API_KEY, COMMOS_LLM_PROVIDER, COMMOS_LLM_MODEL or COMMOS_LLM_BASE_URL for thread assistance. Remove them from the deployment when convenient.

Migration guide

  • Adopt the explicit public overview composition

    Replace the private unreleased Taketide rendering import with brand data and the neutral web screen.

    1. Replace bundle.taketide/public/tracking-preview with bundle.taketide/composition and the explicit creator-platform PublicOverview.web screen.
  • Apply CommunicationOS migrations 0015 and 0016

    Required

    Run the CommunicationOS migration ledger to 0016_tellshift_sequences_rls. applyRls=false applies 0015 and skips 0016, including the scheduler role and the due_sequence_tenants function.

    When: When a deployment runs CommunicationOS persistence.

  • Keep the current authority when reopening bound team runs

    Required

    Removing a source cannot unbind a persisted authority-bound run.

    When: A host reopens a bound team run without its delegation source.

    1. Pass the existing current host authority or resolver to createTeamRun.
    2. Revoke tools or grants through a reduced authority; do not remove the source to bypass binding.
  • Consume the existing bounded worker launch metadata

    Concrete queue and workflow handlers receive the existing delegation and work-order fields.

    When: A worker handler consumes concrete execution-lane backend payloads.

    1. Preserve the delegation projection, workOrderId and deadlineAt through the worker boundary.
    2. Enforce delegated scope and deadlines before effects; retain host-owned durable retry and deduplication handling.
  • Wire the SMS consent gate before enabling outbound SMS

    Required

    createCommunicationDeliveryDispatcher refuses SMS without an smsConsent gate. Back the gate with resolveCommunicationSmsOptOutV1 over the scoped journal.

    When: Before enabling outbound SMS through the CommunicationOS delivery dispatcher.

    1. Apply inbound keywords with toSmsKeywordSignal and applyCommunicationSmsKeywordV1.
    2. Record explicit opt-ins with reviseCommunicationRouteConsentV1 and a consent_capture cause.
    3. Pass an smsConsent gate that resolves the sms_opt_out evidence for the delivery target.
    4. Feed communicationSmsOptOutCheckV1(evidence) into the SMS route eligibility evidence.
  • Treat sms_consent_unknown and sms_opted_out refusals as terminal

    Required

    The dispatcher refuses a non-consented SMS before the provider call with a non-retryable CommunicationDeliveryError (sms_consent_unknown or sms_opted_out, with the gate's evidence refs). A host that retries it, or sends SMS around the dispatcher, reopens the compliance defect this release closes.

    When: Before upgrading a host that sends SMS through createCommunicationDeliveryDispatcher.

    1. Pass the refusal's code and retryable flag (false) to failCommunicationOutboxDelivery, so the outbox record moves to failed on its first attempt; never re-enqueue it.
    2. Route sms_consent_unknown to consent capture, and record a consent_capture revision with reviseCommunicationRouteConsentV1 before any new send.
    3. Keep sms_opted_out final until an inbound START keyword records an opted_in revision; never resend on an operator retry.
    4. If the SMS adapter's consent option is also set, map SmsConsentRefusedError (same codes, retryable false) to CommunicationDeliveryError(code, false), because the dispatcher rethrows adapter errors unchanged.
    5. Never satisfy the gate with a default opted_in state; resolve it from resolveCommunicationSmsOptOutV1 evidence for the delivery target.
  • Save evidence bundles to the lane store before attaching them to a team run

    Required

    Behavioural tightening. attachTaskEvidence and attachRunEvidence now refuse an id that is not a bundle recorded for the same run, completeTask needs every attached id to resolve and one fresh bundle even without requiredEvidence, and finalization refuses run or completed-task ids missing from the store (including ids attached before the upgrade).

    When: When a host attaches team evidence ids that it never saved with store.saveEvidence (or the lane runtime evidence API).

    1. Save each EvidenceBundleRef for the team run (same runId) before calling attachTaskEvidence or attachRunEvidence.
    2. For an in-flight run whose state already references unsaved ids, save the matching bundles before completing its tasks or finalizing it.
  • Enable the verification lane when creating any team run

    Required

    Behavioural tightening. validateTeamRunSpec (and createTeamRun.create) now reject verificationLane.required false for every new run, not only code-producing ones. Persisted runs stay operable: worker add/remove validates in amend mode, and such a legacy run still finalizes without a verifier.

    When: When a host creates team runs with verificationLane.required false.

    1. Set verificationLane.required to true and staff an independent verifier (or a human ownerRole) for the run.
    2. Record the verifier approval with a decidedBy actor that did no task work in the run.
  • Hint tasks for the role that should claim them

    Behavioural tightening. claimNextTask only returns tasks hinted for the worker's role or tasks without a roleHint. A worker no longer falls back to a ready task hinted for another role, so a run whose only worker of a hinted role is absent waits instead of letting another role take the task.

    When: When a run relied on a worker of one role claiming tasks hinted for another role.

    1. Staff a worker for every roleHint the backlog uses, or drop the roleHint on tasks any role may take.
  • Optionally bind a host delegation authority to team runs

    Additive. Without an authority, workers derive from a run-local lead, receive no grant and actor ids stay informational. With one, workers receive only role-required grants the lead holds, creation and start fail closed on a missing tool, grant or depth, and worker-scoped operations must be made by the worker's derived principal.

    When: When a host runs team workers on behalf of a lead principal with Connect tools and grants.

    1. Pass createTeamRun(store, adapter, { delegation }) with the lead principal, its tools and grants, the grants each role requires (roleGrantRequirements), the lead depth and the worker wall-clock bound, or a resolver that returns them per run.
    2. Pass each worker's derivedPrincipalRef (getWorkerDelegation) as actorId to claimNextTask, renewLease, attachTaskEvidence, completeTask, heartbeat and requestHelp.
  • Make team backends idempotent per run and worker

    Additive for backends. startWorker receives delegation, workOrderId and deadlineAt; stopWorker, pauseWorker, resumeWorker and nudgeWorker receive an optional context with the work-order id and deadline, for correlation. A backend call that outlives its deadline is settled indeterminate and may still have run; the worker is then marked failed and the next start issues a new order with a new id, so a backend cannot deduplicate by workOrderId. start now throws TeamWorkOrderFailureError (after recording the failed worker) when a start call fails, instead of throwing the backend error before persisting anything; recorded failures carry only the error class name (and a short code), never the backend message.

    When: When a team backend can apply the same lifecycle call twice, or a caller branches on errors from start, upsertWorker, removeWorker, pause, resume or nudge.

    1. Make startWorker, stopWorker, pauseWorker, resumeWorker and nudgeWorker idempotent by (runId, workerId) and the worker's current state (for example, startWorker adopts a worker that is already running), and forward the delegation projection to the worker instead of broader context.
    2. Catch TeamWorkOrderFailureError and read its orders. reconcileWorkOrders only settles orders past their deadline; check the backend's own state for an indeterminate worker before calling start again.
    3. Expose a stable error name or a short code on backend errors if callers need to tell failures apart; messages are not persisted.
  • Keep team task leases positive and within the worker wall clock

    Required

    Behavioural tightening. claimNextTask and renewLease reject a non-positive lease and a lease longer than the worker's delegated wall clock, which is maxWorkerWallClockSeconds under a delegation authority and one hour (DEFAULT_TEAM_WORKER_TIMEOUT_SECONDS) without one.

    When: When a host claims or renews team task leases longer than one hour, or passes zero or a negative lease.

    1. Renew long-running leases with renewLease before they expire instead of claiming one lease longer than one hour.
    2. Or supply a delegation authority whose maxWorkerWallClockSeconds covers the lease the host needs.
  • Answer team help as the lead, or through a help-response authority

    respondToHelp applies the agent-protocol help response rules. The selected option must be one of the request's safeOptionRefs, the response createdAt must fall between the request's createdAt and expiresAt, and no worker of the run (by worker id or derived principal) may answer. Under a delegation authority, the responder must be the lead the request was routed to, unless createTeamRun receives a helpResponseAuthority that authorizes the response against its evidence. A delegation resolver that returns no authority for a run created under one now fails closed.

    When: When an operator other than the lead answers team help under a delegation authority, or a per-run delegation resolver returns undefined to revoke access.

    1. Pass createTeamRun(store, adapter, { helpResponseAuthority }) and give non-lead responses evidence the authority can check.
    2. Revoke delegation by returning an authority with fewer toolRefs or grantRefs instead of undefined.
  • Enable Gateway-backed thread assistance on communicationos-api

    Additive. Assistance stays off, and endpoints answer 503 ai_unavailable, until an operator opts in.

    1. Ensure the Vercel project can use the AI Gateway, through Vercel OIDC or an AI_GATEWAY_API_KEY project variable.
    2. Set COMMUNICATION_OS_AI_ASSIST_ENABLED=true. Optionally set COMMUNICATION_OS_AI_DEFAULT_MODEL to a catalog id, then redeploy.
  • Return CompanyOS provider OAuth to the monolith

    Required

    The CompanyOS UI now asks the API to return to /companyos/communication-os/connections. The API only redirects there when COMMUNICATION_OS_OAUTH_HOST_ORIGINS maps the CompanyOS session issuer to the CompanyOS origin; otherwise callbacks keep landing on COMMUNICATION_OS_WEB_ORIGIN.

    1. Set COMMUNICATION_OS_OAUTH_HOST_ORIGINS on the CommunicationOS API to a JSON map of the CompanyOS host-exchange issuer to https://www.companyos.run.
    2. Register https://api.hcircle.app/v2/oauth/<network>/callback redirect URIs (email, slack, teams) in each provider app.
    3. Deploy the CommunicationOS API before the web app so /v2/audit/export and providerKey are available.

Upgrade steps

  • Verify SMS consent normalization and fail-closed delivery

    auto

    Run the SMS integration, communication runtime and communication spec checks.

    Packages: @lssm-tech/lib.communication-spec, @lssm-tech/lib.communication-runtime, @lssm-tech/integration.communication-channel-sms

    1. Run bun test in packages/integrations/communication-channel-sms.
    2. Run bun test src/sms-consent.test.ts src/sms-consent-scopes.test.ts src/sms-consent-dispatch.test.ts src/delivery-provider-eligibility.test.ts in the communication runtime.
    3. Run bun run typecheck in the communication spec and runtime.
  • Verify team delegation, help, work orders and readiness

    auto

    Run the execution-lanes tests and typecheck, and the integration.runtime team tests.

    Packages: @lssm-tech/lib.execution-lanes

    1. Run bun test packages/platform/agents/execution-lanes/src (team-run-delegation, team-run-help, team-run-help-response, team-run-work-orders, team-run-evidence-store and runtime/team-readiness cover the new rules).
    2. Run bun test in packages/integrations/runtime for src/channel/execution-lanes-postgres-team-store.pglite.test.ts and the other execution-lanes channel tests.
  • Enable and qualify the hCircle meeting host

    manual

    Configure, deploy the exact commit, then record canary evidence before relying on hCircle meetings.

    Packages: @lssm-tech/lib.communication-spec, @lssm-tech/lib.communication-runtime, @lssm-tech/integration.communication-os-client, @lssm-tech/module.communication-os, @lssm-tech/app.communicationos-api

    1. Create the opa_app and opa_migrator roles in the hCircle governed-data database.
    2. Set HCIRCLE_GOVERNED_DATA_DATABASE_URL (opa_app), HCIRCLE_GOVERNED_RECOVERY_TOKEN, HCIRCLE_GOVERNED_DATA_MIGRATION_URL (opa_migrator), HCIRCLE_MEETING_BYOK_MASTER_KEY and OPA_MEETING_ACCESS_GRANT_PEPPER, then HCIRCLE_MEETING_RUNTIME_ENABLED=true.
    3. Deploy and check that /v2/health reports meetingHost enabled and the deployed commit as release.sha (set COMMUNICATION_OS_BUILD_SHA when VERCEL_GIT_COMMIT_SHA is absent).
    4. Run bun run --cwd packages/apps/communicationos-api canary:meeting-host -- --origin https://api.hcircle.app --sha <commit> --out <evidence.json> and attach the evidence.

Unique release changes

  • Additive. Assistance stays off, and endpoints answer 503 ai_unavailable, until an operator opts in.

    5 packages · 5 occurrences

  • CommunicationOS inbox threads can carry advisory AI triage labels (category, urgency, needs reply, reason) produced through the Vercel AI Gateway. Labels are cached per thread revision, journaled with their model and shown as marked AI chips with filters in the CompanyOS inbox.

    5 packages · 5 occurrences

  • CommunicationOS provider connections, room bindings, governance policies, contact identities and audit export inside CompanyOS

    5 packages · 5 occurrences

  • CommunicationOS thread summaries and AI reply drafts run only through the Vercel AI Gateway. Users can switch between allowlisted Anthropic, OpenAI and Google models per request, and governance managers set the workspace default.

    5 packages · 5 occurrences

  • communicationos-api no longer uses MISTRAL_API_KEY, ANTHROPIC_API_KEY, COMMOS_LLM_PROVIDER, COMMOS_LLM_MODEL or COMMOS_LLM_BASE_URL for thread assistance. Remove them from the deployment when convenient.

    5 packages · 5 occurrences

  • Configure, deploy the exact commit, then record canary evidence before relying on hCircle meetings.

    5 packages · 5 occurrences

  • Meetings join CommunicationOS in CompanyOS: an OPA meeting gets one CommunicationOS conversation with a shared agenda, participants, shared notes and governed follow-ups. The hCircle meeting host becomes production-qualifiable through a startup configuration verdict, a health field, a proven schema ledger and an exact-SHA canary.

    5 packages · 5 occurrences

  • The CompanyOS UI now asks the API to return to /companyos/communication-os/connections. The API only redirects there when COMMUNICATION_OS_OAUTH_HOST_ORIGINS maps the CompanyOS session issuer to the CompanyOS origin; otherwise callbacks keep landing on COMMUNICATION_OS_WEB_ORIGIN.

    5 packages · 5 occurrences

  • Run the CommunicationOS migration ledger to 0016_tellshift_sequences_rls. applyRls=false applies 0015 and skips 0016, including the scheduler role and the due_sequence_tenants function.

    4 packages · 4 occurrences

  • Tellshift sequences, consent-bound email enrollment, durable due discovery, approved Gmail send and tenant erasure on CommunicationOS

    4 packages · 4 occurrences

  • Communication spec drops its ProviderSpec dependency and two integrations publish caret workspace ranges; no API changes.

    3 packages · 3 occurrences

  • createCommunicationDeliveryDispatcher refuses SMS without an smsConsent gate. Back the gate with resolveCommunicationSmsOptOutV1 over the scoped journal.

    3 packages · 3 occurrences

  • Run the SMS integration, communication runtime and communication spec checks.

    3 packages · 3 occurrences

  • SMS STOP/START/HELP keywords become CAS preference revisions with sms_opt_out evidence, and outbound SMS fails closed without durable consent

    3 packages · 3 occurrences

  • The dispatcher refuses a non-consented SMS before the provider call with a non-retryable CommunicationDeliveryError (sms_consent_unknown or sms_opted_out, with the gate's evidence refs). A host that retries it, or sends SMS around the dispatcher, reopens the compliance defect this release closes.

    3 packages · 3 occurrences

  • Concrete queue and workflow handlers receive the existing delegation and work-order fields.

    2 packages · 2 occurrences

  • Keep public overview rendering in the neutral creator-platform experience and brand content in Taketide composition.

    2 packages · 2 occurrences

  • Preserve persisted team authority and the exact bounded projection through concrete backends

    2 packages · 2 occurrences

  • Removing a source cannot unbind a persisted authority-bound run.

    2 packages · 2 occurrences

  • Replace the private unreleased Taketide rendering import with brand data and the neutral web screen.

    2 packages · 2 occurrences

  • Additive for backends. startWorker receives delegation, workOrderId and deadlineAt; stopWorker, pauseWorker, resumeWorker and nudgeWorker receive an optional context with the work-order id and deadline, for correlation. A backend call that outlives its deadline is settled indeterminate and may still have run; the worker is then marked failed and the next start issues a new order with a new id, so a backend cannot deduplicate by workOrderId. start now throws TeamWorkOrderFailureError (after recording the failed worker) when a start call fails, instead of throwing the backend error before persisting anything; recorded failures carry only the error class name (and a short code), never the backend message.

    1 package · 1 occurrence

  • Additive. Without an authority, workers derive from a run-local lead, receive no grant and actor ids stay informational. With one, workers receive only role-required grants the lead holds, creation and start fail closed on a missing tool, grant or depth, and worker-scoped operations must be made by the worker's derived principal.

    1 package · 1 occurrence

  • Behavioural tightening. attachTaskEvidence and attachRunEvidence now refuse an id that is not a bundle recorded for the same run, completeTask needs every attached id to resolve and one fresh bundle even without requiredEvidence, and finalization refuses run or completed-task ids missing from the store (including ids attached before the upgrade).

    1 package · 1 occurrence

  • Behavioural tightening. claimNextTask and renewLease reject a non-positive lease and a lease longer than the worker's delegated wall clock, which is maxWorkerWallClockSeconds under a delegation authority and one hour (DEFAULT_TEAM_WORKER_TIMEOUT_SECONDS) without one.

    1 package · 1 occurrence

  • Behavioural tightening. claimNextTask only returns tasks hinted for the worker's role or tasks without a roleHint. A worker no longer falls back to a ready task hinted for another role, so a run whose only worker of a hinted role is absent waits instead of letting another role take the task.

    1 package · 1 occurrence

  • Behavioural tightening. validateTeamRunSpec (and createTeamRun.create) now reject verificationLane.required false for every new run, not only code-producing ones. Persisted runs stay operable: worker add/remove validates in amend mode, and such a legacy run still finalizes without a verifier.

    1 package · 1 occurrence

  • Communication projections use the canonical pure help definition

    1 package · 1 occurrence

  • respondToHelp applies the agent-protocol help response rules. The selected option must be one of the request's safeOptionRefs, the response createdAt must fall between the request's createdAt and expiresAt, and no worker of the run (by worker id or derived principal) may answer. Under a delegation authority, the responder must be the lead the request was routed to, unless createTeamRun receives a helpResponseAuthority that authorizes the response against its evidence. A delegation resolver that returns no authority for a run created under one now fails closed.

    1 package · 1 occurrence

  • Run the execution-lanes tests and typecheck, and the integration.runtime team tests.

    1 package · 1 occurrence

  • Team workers run under bounded, least-privileged delegation with typed help and two-phase lifecycle work orders, and team readiness and evidence follow the finalization rules

    1 package · 1 occurrence

Impacted packages

  • @lssm-tech/app.communicationos-api

    Layer: apps · 2 changes

  • @lssm-tech/app.taketide-web

    Layer: apps · 2 changes

  • @lssm-tech/bundle.managed-companyos

    Layer: bundles · 2 changes

  • @lssm-tech/bundle.taketide

    Layer: bundles · 2 changes

  • @lssm-tech/integration.communication-channel-email

    Layer: integrations · 2 changes

  • @lssm-tech/integration.communication-channel-sms

    Layer: integrations · 4 changes

  • @lssm-tech/integration.communication-os-agent

    Layer: integrations · 1 change

  • @lssm-tech/integration.communication-os-client

    Layer: integrations · 2 changes

  • @lssm-tech/integration.communication-os-persistence

    Layer: integrations · 2 changes

  • @lssm-tech/integration.communication-os-runtime

    Layer: integrations · 2 changes

  • @lssm-tech/integration.provider-linkedin

    Layer: integrations · 1 change

  • @lssm-tech/integration.runtime

    Layer: integrations · 3 changes

  • @lssm-tech/integration.runtime-local

    Layer: integrations · 1 change

  • @lssm-tech/lib.communication-runtime

    Layer: libs · 2 changes

  • @lssm-tech/lib.communication-spec

    Layer: libs · 1 change

  • @lssm-tech/lib.execution-lanes

    Layer: libs · 9 changes

  • @lssm-tech/module.communication-os

    Layer: modules · 2 changes