Channel ingress API

Channel ingress is the experimental access-control boundary for inbound channel events. Plugins own platform facts and side effects; core owns generic policy: DM/group allowlists, pairing-store DM entries, route gates, command gates, event auth, mention activation, redacted diagnostics, and admission.

Use openclaw/plugin-sdk/channel-ingress-runtime for receive paths.

Runtime resolver

import {
  defineStableChannelIngressIdentity,
  resolveChannelMessageIngress,
} from "openclaw/plugin-sdk/channel-ingress-runtime";

const identity = defineStableChannelIngressIdentity({
  key: "platform-user-id",
  normalize: normalizePlatformUserId,
  sensitivity: "pii",
});

const result = await resolveChannelMessageIngress({
  channelId: "my-channel",
  accountId,
  identity,
  subject: { stableId: platformUserId },
  conversation: { kind: isGroup ? "group" : "direct", id: conversationId },
  contextBinding: {
    agentId: agentRoute.agentId,
    sessionKey: agentRoute.sessionKey,
    messageId,
    inboundEventKind: "user_request",
  },
  event: { kind: "message", authMode: "inbound", mayPair: !isGroup },
  policy: {
    dmPolicy: config.dmPolicy,
    groupPolicy: config.groupPolicy,
    groupAllowFromFallbackToAllowFrom: true,
  },
  allowFrom: config.allowFrom,
  groupAllowFrom: config.groupAllowFrom,
  accessGroups: cfg.accessGroups,
  route,
  readStoreAllowFrom,
  command: hasControlCommand ? { allowTextCommands: true, hasControlCommand } : undefined,
});

const ctx = runtime.channel.inbound.buildContext({
  // Pass the exact host result; do not rebuild participant evidence from
  // SenderId, From, session keys, routes, rooms, or message metadata.
  channelIngress: result,
  // ...normalized channel facts
});

Do not precompute effective allowlists, command owners, or command groups. The resolver derives them from raw allowlists, store callbacks, route descriptors, access groups, policy, and conversation kind.

For a result that will enter a host context, resolve after the channel’s route owner has selected the final agent and session. contextBinding freezes those facts with the stable transport message id (when present) and final inbound event kind. Decision-only checks may omit it, but such a result is not valid execution provenance and must not be passed as channelIngress. When a channel batches several admitted messages, pass their exact results in source order; the finalized context message id identifies the last source result.

Product participant identity

An identity descriptor may provide resolveParticipant(subject), returning { domain, idKind, id } only when the plugin can prove those remote facts. The domain belongs to the remote service: for example, a Slack workspace or an application-scoped identity issuer. It is not OpenClaw’s local accountId. Keep user IDs, bot IDs, and proxy identities distinct when the service gives them different meanings. Names and successful Gateway profile lookups are not identity evidence.

Pass the exact resolver result through the host-injected context builder. The host carries product facts privately until accepted input is recorded in the session participant aggregate. Raw product identity is not added to the public diagnostic result. An unqualified producer retains an unresolved observation instead of becoming a Gateway profile or an invented remote subject. Product participation does not grant access.

This product path is independent of execution-identity audit collection: it neither enables that collection nor reuses its HMAC references or opaque diagnostic carriers. Audit’s separate evidence contract is described below.

Result

Bundled plugins should consume modern projections directly:

Field	Meaning
ingress	ordered gate decision and admission
senderAccess	sender/conversation authorization only
routeAccess	route and route-sender projection
commandAccess	command authorization; requested: false when no command gate ran
activationAccess	mention/activation result

Event authorization stays available on the ordered ingress.graph and the decisive ingress.reasonCode; no separate event projection is emitted.

Deprecated third-party SDK helpers may rebuild older shapes internally. New bundled receive paths should not translate modern results back into local DTOs.

When execution-identity audit collection is enabled, a trusted active native plugin is the authoritative in-process producer of its remote participant fact. The host-injected registered runtime binds the resolver result to the exact plugin record and registry lifecycle epoch, then validates its complete available conversation, route, agent, session, message, event, and participant scope during a one-shot context handoff. The public standalone builder remains non-authoritative and cannot mint participant evidence. Queue collection retains attribution only when every contribution has valid evidence for the same participant; mixed, missing, stale, or unminted evidence is unknown. The carrier is opaque, bounded, one-shot, and diagnostic only. Plugins cannot mint participant evidence from caller-chosen sender, account, room, route, session, message, or transport fields. The SDK intentionally exposes no record, epoch, owner capability, participant-evidence constructor, or evidence copier. A structurally similar result, stale record, reused result, or scope-changed context does not gain host authority.

boundary-verified means core verified that the participant fact crossed this trusted active registered native-plugin boundary with the exact record, epoch, scope, and one-shot handoff. It does not mean core independently queried the remote service; only the channel plugin can observe that transport fact.

The audit states are distinct:

• supported: the authoritative ingress resolver ran. Its exact result can yield a present invoker and enforced or attribution-only coverage.
• unknown: a supported handoff was missing, stale, fake, reused, mixed, or otherwise failed host validation. Unknown never means allowed.
• unsupported: a named path has no Phase 0 authoritative integration and explicitly passes channelIngress: "unsupported". Unsupported never means allowed and is not a shortcut for incomplete wiring.

Identifier authentication

IdentifierAuthentication grades an identifier claim as verified, asserted, unverified, or mutable, strongest to weakest. It is an input to channel authorization only. It is not a principal, grant, relationship, or execution-identity assurance strength. In particular, an identifier claim of verified never becomes execution assurance boundary-verified or cryptographic.

Import the type and meetsIdentifierAuthentication(actual, minimum) from openclaw/plugin-sdk/channel-ingress-runtime. Downstream authentication mappers should use this boolean comparator instead of maintaining their own rank tables.

The meanings are normative:

• verified: the owning trusted transport or session boundary bound this exact identifier to this sender.
• asserted: a trusted boundary vouched for the sender without binding this exact identifier.
• unverified: the identifier is exact and stable, but claimed ownership was not proved.
• mutable: the identifier is a changeable or shared alias, such as a display name.

Declare verified only from transport or session metadata controlled by the owning boundary. Sender-controlled content, model input, ordinary message context, routing metadata, and the integrity of the host admission carrier do not establish it.

The kernel preserves the exact redacted allowlist-entry to subject-identifier pair that matched. It combines the entry and subject claims by taking the weaker claim for that exact pair, then compares it with minIdentifierAuthentication. Identifiers of the same kind remain distinct, so a weak secondary email does not weaken a separately matched verified email.

A subject that supplies a per-message authentication map must claim every field it wants counted. A field missing from a supplied map is treated as unverified, even if its identity descriptor declares a stronger static claim. Channels with static strength omit the map entirely.

Expose classifyEntryAuthentication: identityEntryAuthenticationClassifier(identity) from the security adapter’s resolveDmPolicy result, importing the helper from openclaw/plugin-sdk/channel-ingress-runtime. It uses the identity descriptor’s entry normalizers and returns the strongest static claim among accepting fields, or undefined when none accepts the entry; wildcard entries are excluded. The security audit counts configured allowFrom entries that depend only on mutable identifiers: it warns when name matching is disabled and, when enabled, previews how many entries would stop authorizing after disabling it. Findings contain counts and config paths, not raw entries; pairing-store approvals are outside this check. Symbolic accessGroup: references resolve membership separately and are not counted as mutable identifiers.

Existing plugins remain source-compatible during the deprecation window:

Deprecated field	Exact mapping
dangerous: true	authentication: "mutable"
dangerous: false or omitted	default authentication: "asserted"
mutableIdentifierMatching: "enabled"	minimum mutable
mutableIdentifierMatching: "disabled" or omitted	default minimum asserted

An explicit authentication or minIdentifierAuthentication takes precedence. The deprecated fields remain through the current Plugin SDK major and are planned for removal in the next major after bundled and known external plugins migrate.

Bundled channel declarations

Bundled channels use the strongest claim supported by every receive path that shares an identity declaration. These are channel-authorization claims, not execution-identity assurance:

Channel	Identifier claim	Authoritative transport or session fact
Discord	Gateway user ID: verified; PluralKit member ID: asserted; names and tags: mutable	Discord supplies author.id or user.id on events delivered over the authenticated bot-token Gateway session. PluralKit member IDs come from its authenticated API response, not the Discord Gateway.
Google Chat	sender.name: verified; email: mutable	The webhook validates Google’s signed token, issuer, and configured audience before consuming the Google-owned event body.
IRC	server connection prefix and user@host: asserted; nick-based aliases: mutable	The selected IRC server vouches for the connection prefix, but the generic transport does not prove account ownership.
Mattermost	post user ID: verified; username: mutable	The authenticated Mattermost WebSocket emits server-owned post events whose post.user_id identifies the author.
Microsoft Teams	sender and conversation IDs: asserted; sender name: mutable	Bot Framework authenticates the connector activity, but the plugin does not independently prove exact ownership of every ID representation.
Slack	user and workspace-user IDs: asserted; names and slugs: mutable	Direct Slack delivery binds user IDs, while relay mode authenticates the relay peer without an end-to-end exact-sender attestation. The shared declaration uses the defensible common claim.

If a receive path cannot support the declaration shared by its channel, split the declaration or supply a weaker per-message claim. Never infer a stronger claim from message text, routing, or host evidence-carrier integrity.

Access groups

accessGroup:<name> entries stay redacted. Core resolves static message.senders groups itself and calls resolveAccessGroupMembership only for dynamic groups that require a platform lookup. Missing, unsupported, and failed groups fail closed.

Event modes

authMode	Meaning
inbound	normal inbound sender gates
command	command gates for callbacks or scoped buttons
origin-subject	actor must match the original message subject
route-only	route gates only for route-scoped trusted events
none	plugin-owned internal events bypass shared auth

Use mayPair: false for reactions, buttons, callbacks, and native commands.

Routes and activation

Use route descriptors for room, topic, guild, thread, or nested route policy:

route: {
  id: "room",
  allowed: roomAllowed,
  enabled: roomEnabled,
  senderPolicy: "replace",
  senderAllowFrom: roomAllowFrom,
  blockReason: "room_sender_not_allowlisted",
}

Use channelIngressRoutes(...) when a plugin has several optional route descriptors; it filters disabled branches while keeping route facts generic and ordered by each descriptor’s precedence.

Mention gating is an activation gate. A mention miss returns admission: "skip" so the turn kernel does not process an observe-only turn. Most channels should leave activation after sender and command gates. Public chat surfaces that must quiet non-mentioned traffic before sender allowlist noise can opt into activation.order: "before-sender" when text-command bypass is disabled. Channels with implicit activation, such as replies in bot threads, resolve channels.defaults.implicitMentions plus channel and account overrides with resolveChannelImplicitMentions(...), then pass the result as activation.implicitMentions. The projected activationAccess.shouldBypassMention reports when command or implicit activation bypassed an explicit mention.

Redaction

Raw sender values and raw allowlist entries are resolver input only. They must not appear in resolved state, decisions, diagnostics, snapshots, or compatibility facts. Use opaque subject ids, entry ids, route ids, and diagnostic ids.

Verification

pnpm test src/channels/message-access/message-access.test.ts src/plugin-sdk/channel-ingress-runtime.test.ts
pnpm plugin-sdk:api:diff --base "$(git merge-base origin/main HEAD)" --head HEAD