* feat(logbook): automatic work journal plugin with a plugin-contributed Control UI tab
Squash of PR #99930 work for rebase onto the Control UI route refactor:
- extensions/logbook: Dayflow-style capture -> observations -> timeline cards
pipeline with SQLite store, node capture commands, standup/ask, retention
- plugin SDK/gateway seam: surface "tab" Control UI descriptors projected
into hello-ok controlUiTabs (scope-filtered, deterministic order)
- Control UI: dynamic plugin tabs with bundled Logbook view
- docs, tests, labeler wiring
* feat(ui): port plugin tabs and Logbook to the route-owned Control UI architecture
- shared /plugin route carries the tab id in the query (?id=<tab>), matching
the router's exact-path contract
- openclaw-plugin-page renders bundled views (Logbook), sandboxed plugin
frames (descriptor path), or the unavailable card
- sidebar renders hello controlUiTabs after each group's static routes
- Logbook view/controller live under ui/src/pages/plugin/
* fix(ui): namespace plugin tabs by pluginId to prevent cross-plugin tab id collisions
* fix(logbook): prefer app capture nodes and rotate off failing nodes
* fix(plugins): reject protocol-relative Control UI tab paths
* fix(logbook): harden automatic journal
* docs(changelog): remove maintainer self-credit
* chore(ui): refresh locale metadata after rebase
* fix(logbook): preserve analysis window boundaries
* fix(logbook): align status privacy and timezone
* fix(ui): stop hidden plugin tab polling
Wires replyOptions.onModelSelected so the resolved provider/model and
thinking level for each turn (including after fallback) are sent as
author_model / author_thinking fields on activity rows and the final
reply. Servers without these columns ignore the unknown JSON fields, so
the wire shape is backward compatible; servers that persist them get
per-message attribution.
- docs: document agentActivity option, the non-inherited agent_activity:write
scope, default-off and best-effort degradation behavior
- activity: type the enqueue catch parameter as unknown (lint lane)
- activity: treat toolCallId as opaque; normalize only itemId lane prefixes
- activity: skip identical/stale-shorter commentary snapshots so no
redundant PATCHes queue
- http-client: fail fast when createActivityMessage has no channel or
conversation target; narrow mocked request bodies before JSON.parse
* feat(control-ui): session grouping with drag & drop and channel categorization
* fix(control-ui): restrict session drag&drop to private session-key payloads
* chore(i18n): translate session grouping strings across control ui locales
* chore(protocol): regenerate swift gateway models for session category
* chore(protocol): regenerate swift gateway models for session category
* fix(imessage): only warn about empty group allowlist when messages actually drop
With groupPolicy="allowlist", a non-empty effective groupAllowFrom admits
group messages even when channels.imessage.groups is empty (senderFilterBypass
in src/config/group-policy.ts), so the startup warning "Every inbound group
message will be dropped" fired as a false positive for that configuration.
The warning now mirrors the runtime gate's effective sender allowlist (same
allowFrom fallback + legacy chat-target merge) and fires only when both the
groups registry and the effective group sender allowlist are empty - the only
startup-provable drop-all config. The message now names groupAllowFrom as the
fix, since adding groups entries alone leaves the sender gate blocking.
Docs: describe the two group gates' warnings separately with per-warning
remedies instead of implying both fire for the same config.
* docs(imessage): format migration guide
* fix(imessage): warn on empty group sender allowlist
* docs: move iMessage fix to unreleased
Externalized provider plugins (qwen, moonshot, zai, deepseek, groq, cerebras, chutes) are excluded from dist packaging, so their manifest-declared endpoint classes were invisible to installed gateways and built source checkouts: DashScope/Moonshot-class base URLs classified as custom, breaking image prompt placement, streaming-usage and developer-role compat defaults, and deterministically failing image.test.ts / model-compat DashScope cases after pnpm build. The official external provider catalog now mirrors each plugin's providerEndpoints and provider attribution appends catalog metadata after installed/bundled manifests (first match wins; repo-bundled catalog only, hosted feeds never influence classification). Includes a catalog-manifest mirror contract test and a dist-simulation regression test.
CLI harnesses (claude-cli, gemini-cli) cannot enforce runtime toolsAllow, so
Crestodian previously fell back to the single-turn planner for them. The
openclaw-tools stdio MCP entry can now serve the ring-zero crestodian tool
(OPENCLAW_TOOLS_MCP_TOOLS=crestodian), and Crestodian CLI runs inject it as the
run's exclusive MCP surface: no loopback server, no plugin/user MCP servers,
with the server kept under the "openclaw" name so existing tool pre-approvals
apply. agent-turn routes configured or detected CLI-harness models through
runCliAgent with native session resume across turns, keeps embedded toolsAllow
enforcement for API-key models and the Codex app-server fallback, and still
degrades to the planner when no loop backend works.
The host-verified approval contract survives the process boundary: per-turn
arming and the pending exact-operation hash travel to the stdio server via the
generated MCP config env, and the host mirrors proposal transitions back from
harness tool events (denial registers the hash, mismatch voids it, execution
consumes it) so a generic yes can never authorize a different mutation. The
trust model (the CLI owns its native tools) is documented in
docs/cli/crestodian.md.
* feat(crestodian): AI-first conversational onboarding with typed-op guardrails
Interactive `openclaw onboard` (and bare `openclaw` on a fresh install) now
opens the Crestodian conversation: detection-backed first-run proposal
(Claude Code/Codex logins, API keys), persona AI turns for every free-form
message (configless local-runtime fallback, 60s deadline, deterministic
degradation), approval-gated typed operations, chat-hosted channel setup
(`connect <channel>`), config get/schema read ops with secret redaction, and
a post-write validation hook that feeds schema errors back for a self-fix
turn. Adds the additive gateway `crestodian.chat` method so app clients run
the same conversation. Classic wizard stays behind --classic/explicit flags;
non-interactive automation unchanged; `--modern` becomes a deprecated alias
for `openclaw crestodian`.
* feat(macos): Crestodian chat onboarding and importance-ordered permissions
Replace the gateway step-wizard page with a Crestodian chat over the new
crestodian.chat method (works before any model auth exists), sort the
permissions page by importance with no scrolling, drop the redundant manual
refresh, and bump the onboarding version.
* feat(crestodian): run the custodian on the real agent loop with a ring-zero tool
Crestodian conversations now execute through the same embedded agent runner
as regular agents: a persistent agent session with a single construction-gated
`crestodian` tool wrapping the typed operations (read actions free; mutations
require approved=true asserted from explicit user consent, audited, with
post-write config validation fed back into the loop). The engine prefers the
loop (configured models or the Codex app-server fallback) and degrades to the
single-turn planner, then to deterministic commands. Setup approval seeds the
crestodian exec approval so local model harnesses can run; the configless
Codex backend config now enables exec and direct tool loading (it was
dead-on-arrival behind tools.exec.mode=deny and the tool-search index).
* test(crestodian): type the engine mock signatures for the core test lane
* fix(crestodian): map the advertised create_agent tool action
* fix(crestodian): host-verified approval arming and a macOS setup completion gate
Review findings: the model-supplied approved flag alone could authorize
ring-zero mutations (prompt injection / model error), and removing the macOS
wizard gate let users Next past the Crestodian page with nothing configured.
Mutating tool actions now also require host-verified consent (the engine arms
approval only when the user's actual message is an explicit yes), and local
macOS onboarding blocks advancing until setup authored the config, using the
same signal the old step wizard checked.
* fix(crestodian): bind approval to the exact proposed operation and gate dot navigation
A generic yes no longer authorizes arbitrary mutations: denied mutating tool
calls register a canonical operation fingerprint (host-owned, per session),
and an armed turn executes only the identical call, once. The denial message
is arming-aware so the approved turn self-heals in one roundtrip, and the
agent protocol pre-registers proposals. macOS onboarding page dots now honor
the same setup-completion gate as the Next button.
* fix(crestodian): redact sensitive wizard answers, skip logged-out CLIs, gate programmatic advance
Sensitive channel-wizard answers (tokens, passwords) are redacted from the
AI-visible conversation history; setup and the onboarding welcome never pick
or advertise a definitively logged-out CLI as the model; and macOS
handleNext() honors the page gates for programmatic callers (chat handoff)
just like the Next button.
* fix(crestodian): align the onboarding welcome's configured predicate with the app gate
A valid config carrying only a default model (partial/hand-written) now still
gets the first-run proposal instead of the ready guide, so the macOS setup
gate can always be satisfied from the conversation.
* fix(crestodian): armed turns can never mint their own executable proposal
An approval-mismatched call inside an armed turn no longer re-registers and
invites a retry (which let the model swap the approved operation for another
in the same turn); it voids the approval entirely and requires a fresh yes.
Proposals register only in unarmed turns, which the agent protocol already
does when proposing.
* fix(onboard): route any explicit setup flag to the classic wizard
* fix(ci): satisfy new lint rules, tool-display guard, and generated artifacts for crestodian
* chore(i18n): refresh native inventory after permissions copy wrap
* fix(crestodian): harden conversational onboarding
* docs(crestodian): document conversational onboarding
* test(crestodian): type embedded runner mock
* fix(crestodian): close onboarding security gaps
* chore: retrigger ci
* refactor: extract reusable AI runtime package
* refactor: complete AI provider relocation
* refactor: keep llm core internal
* refactor(ai): make @openclaw/ai self-contained with host policy ports
Move pure transport helpers (tool projections, strict-schema normalization,
prompt-cache boundary, stream guards, anthropic/openai compat, request
activity) from src into packages/ai; move utf16-slice into
normalization-core. Inject host policy (guarded fetch, redaction,
strict-tool defaults, diagnostics logging) through AiTransportHost with
inert library defaults installed by src/llm/stream.ts. Narrow the public
barrel to instance-scoped createApiRegistry/createLlmRuntime; the
process-default runtime moves behind internal/ and
registerBuiltInApiProviders takes an explicit registry. Delete the
src/llm/api-registry re-export facade.
* fix(ai): teach node, jiti, and vite resolvers the @openclaw/ai and utf16-slice subpaths
The workspace alias tables in root-alias.cjs, plugin-sdk-native-resolver,
sdk-alias, the shared vitest config, and the Control UI vite config only
knew @openclaw/llm-core; Node-side plugin loading resolved @openclaw/ai
through the pnpm symlink to the unbuilt dist (checks-node-compact CI
failures), and the Control UI build broke on the new
normalization-core/utf16-slice subpath.
* chore(ui): drop leftover service-worker debug logging
* build(release): ship @openclaw/ai with its own shrinkwrap and honest dependency set
packages/ai declares only its six real runtime deps (kysely, chalk, json5,
tslog, zod, fs-safe, and proxyline were never imported); orphaned root deps
removed. generate-npm-shrinkwrap now treats publishable packages/* like
publishable plugins so the AI tarball pins its transitive tree even though
workspace deps are omitted from the root shrinkwrap. knip learns the
package entry points; the tsdown dts neverBundle option moves to its
documented deps.dts home; the README documents the no-semver internal/*
contract and host ports.
* docs(ai): add minimal external-consumer example app
examples/ai-chat consumes only the public @openclaw/ai surface (built dist
via the workspace link): isolated runtime, built-in provider registration,
one streamed completion. Supports Anthropic/OpenAI via env keys and a
keyless local Ollama target; live-verified against Ollama.
* docs(ai): document the @openclaw/ai package and workspace shrinkwrap boundary
* chore(check): include examples/ in duplicate-scan targets
* fix: emit normalization package subpaths
* fix: complete AI package boundary artifacts
* fix: align AI package boundary contracts
* fix(ci): stabilize package release contracts
* test: align documentation contract checks
* test: keep cron docs guard aligned
* test: align restored docs contract guards
* test: follow upstream docs contracts
* docs: drop superseded talk wording
The /set/media control route accepts dark|light|no-preference|none, but the
CLI rejected no-preference, so prefers-color-scheme: no-preference could not
be emulated from the CLI. Align the CLI argument, validation, and error text
with the route contract and update the browser-control docs media line.
Allow busy TUI sessions to forward prompts into the configured queue while
keeping queued-turn admission, cancellation, restart, and transcript ownership
consistent across the TUI, Gateway, and followup queue.
Co-authored-by: Sebastien Tardif <sebtardif@ncf.ca>
The non-interactive permission failure surfaces as acpx's
PermissionPromptUnavailableError, not AcpRuntimeError; acpx 0.11.2 throws
it unwrapped (isGenericInternalAcpErrorMessage only matches "Internal
error"). Matches the corrected fact in the acp-agents-setup rewrite.
Pre-commit hook bypassed with formatting proof: repo oxfmt + Mintlify
accordion repair already run on the file (worktree has no node_modules).
* feat(slack): add ignoreOtherMentions channel config
Mirrors the existing Discord `ignoreOtherMentions` option for Slack
channels. When set on a channel entry, drop channel/group/MPIM messages
that mention another user or subteam but not this bot — the inverse of
`requireMention`. Useful in busy channels where the bot would otherwise
reply to side conversations.
Implementation lives in `prepareSlackMessage` right after the
`allowBots: "mentions"` bot-drop gate and before the `shouldRequireMention`
non-mention drop, so it integrates with the new `messageIngress` ingress
pipeline. The gate is conditional on `canDetectMention` (botUserId
resolvable via `auth.test` or explicit mention regexes configured) to
avoid false drops when we have no reliable way to tell bot vs non-bot
mentions apart.
Slack implicit mentions (thread participation) are intentionally NOT
respected here — they fire for every message in a bot-participated
thread, so honoring them would defeat the feature in any active thread.
The gate matches on `wasMentioned` (explicit) and `hasAnyMention`, with
the existing `shouldBypassMention` override (e.g. authorized commands)
also respected.
Includes config types, zod schema, channel-config resolution,
prepare-message implementation, full test coverage in
`prepare.test.ts`, regenerated channel + docs baselines, and Slack
docs entry.
* refactor(slack): harden other-mention filtering
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
* fix(telegram): resolve local Bot API container file paths against trustedLocalFileRoots
When the self-hosted telegram-bot-api server runs with --local inside a
container, getFile returns absolute file_path values rooted at the
container data dir (/var/lib/telegram-bot-api/...). The host process
mounts that volume at a different path, so the absolute path never
matches a trustedLocalFileRoots entry and inbound media fails to
materialize (messages reach the agent as bare <media:document>
placeholders without bytes).
Map container-absolute paths back to relative candidates and resolve
them under each trusted root, trying both with and without the
per-token directory segment the local server uses. Relative file_path
values are now also resolved against trusted roots before falling back
to the HTTP file endpoint, which keeps large-file (>20MB) local-mode
downloads working.
Path traversal protections: candidates are normalized, NUL bytes and
dot-segments are rejected, and reads stay behind the trusted-roots
file-access sandbox.
AI-assisted (Claude Code), validated with targeted vitest run.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Co-Authored-By: Dizesales <269209351+Dizesales@users.noreply.github.com>
* fix(telegram): map local Bot API container media paths
Co-authored-by: Lucas Magalhaes <ellucasrj@gmail.com>
* chore(config): refresh channel metadata
---------
Co-authored-by: Lucas Magalhaes <ellucasrj@gmail.com>
Co-authored-by: Dizesales <269209351+Dizesales@users.noreply.github.com>
Co-authored-by: Peter Steinberger <steipete@gmail.com>
