* fix(gateway): harden hook admission and shutdown
Flush structured 413 and 408 hook responses before closing oversized or timed-out request sockets, and reject deferred wake requests that target an explicit session.
Direct Gateway close now marks startup and readiness as draining immediately and attempts every HTTP listener shutdown before reporting an incomplete close.
* test(gateway): model request destruction in fuzzing
Give the HTTP request fuzz double the IncomingMessage destroy contract exercised after 413 and 408 responses finish.
* fix(gateway): preserve response and shutdown bounds
Route installed webhook body-limit failures through the shared response-first close owner and require complete Feishu 413/408 responses before terminal connection closure.
Retain the current-main grace bound for agent-harness disposal so one stuck harness cannot block listener teardown.
* test(telegram): model response-close request cleanup
Keep the slow-body request alive through response finish, then assert cleanup when the mock response transport closes.
The `## Delegation` guidance added in #125691 lived only in
buildAgentSystemPrompt, so Codex-runtime agents never received it: the
Codex harness builds its own developer instructions in
extensions/codex/src/app-server/thread-prompt.ts and imports nothing
from the system-prompt builders. Live A/B on gpt-5.6-luna had the native
runtime answer "spawn a visible session" while the Codex runtime
answered "spawn a hidden subagent".
Move the policy into src/agents/delegation-guidance.ts, owning both the
main-session mode resolver and the section text, and export it through
the agent-harness plugin SDK barrel that the Codex harness already uses.
The hidden-delegation vocabulary is injected by each runtime, so core
never names a plugin-owned tool: native passes `sessions_spawn`, Codex
passes native `spawn_agent`. Visible sessions stay `sessions_spawn`
with visible=true on both runtimes because Codex-native children are
never OpenClaw sessions.
Also narrows the Codex line that told the model to use `sessions_spawn`
only for OpenClaw/ACP delegation; it now scopes that to internal
legwork, so user-facing deliverables still route to a visible session.
* fix: honor inherited model pins over channel defaults
Child sessions now keep parent model pins consistently across reply, status, harness prediction, and agentCommand.
* test: align agent command model selection fixtures
Keep concurrent override and threaded parent-selection coverage aligned with canonical model normalization and runtime snapshot access.
* fix(ci): stop codex lane cold-graph hangs
The side-question domain-policy test loaded the complete agent-harness tool graph inside a one-second readiness race, making the serial non-isolated Codex shard fail or stay silent under cold imports. Build the test's web_search marker and real web_fetch tool from the narrow implementation, then synchronize on turn startup before issuing the tool call. Cap each Codex test process at 12 files so CI gets bounded time-to-first-output as defense in depth.\n\nRefs #125839
* fix(test): keep codex web fetch fixture on sdk boundary
Load the real web_fetch factory on demand through the existing local-only plugin test runtime. This preserves the narrow cold-graph fix without letting a bundled plugin test reach into core internals.
Reuse the canonical realtime audio duration and OpenAI-compatible wire-format helpers while removing redundant state transitions and truncate payload construction. The production diff is net-negative.
Provider interruption policy and lifecycle ownership stay local to each provider; this does not introduce shared mutable assistant tracking or change behavior.
* fix(auth): preserve WHAM classifications and failure recording
WHAM 401/403 state now drives accurate re-auth guidance, while inline hook failures are contained after persistence so recorded failures cannot escape or be masked.
* docs(plugin-sdk): define auth cooldown classifications
document the additive cooldown diagnostic contract and cover its canonical public-SDK projection.
* fix(auth): keep WHAM diagnostics source-compatible
keep cooldownReason canonical, persist exact WHAM diagnostics in optional cooldownClassification, and preserve operator guidance plus failure-hook containment.
* fix(auth): keep failover on canonical cooldown reasons
ensure optional WHAM diagnostics never drive scheduling and discard mismatched persisted reason/classification pairs.
* fix(setup): refresh Codex registry with staged install
* fix(macos): verify inference before onboarding handoff
* fix(setup): use native Codex home for subscription auth
* fix(codex): honor attempt-scoped setup config
* fix(macos): align onboarding handoff with reopen
* fix(setup): await prepared model convergence
* fix(ui): avoid false auth state for empty catalog
* fix(setup): scope catalog convergence to Codex gateway
* fix(setup): publish the committed runtime catalog
* fix(models): project configured static runtime models
* fix(codex): expose app-server model catalog
* fix(models): preserve Codex auth across reloads
* fix(ci): align Codex onboarding checks
* test(ui): stabilize dock suppression environment
* fix(codex): honor discovery config in app-server model catalog
The manifest documents discovery.enabled (bundled fallback list) and
discovery.timeoutMs (default 2500ms) for model discovery; the new catalog
path used the generic 60s request timeout and ignored the enable gate.
Also drop the test-only listModels injection seam in favor of vi.mock.
* fix(setup): refuse prepared Codex auth over an explicit remote transport
configureCodexCliPreparedAuth silently rewrote an explicitly configured
websocket/unix app-server to local stdio (keeping a dangling url), moving
the credential boundary onto this host. Fail setup with actionable
guidance instead; also surface the root cause when the prepared model
catalog refresh fails after activation.
* refactor(agents): one canonical model-catalog identity key
Three near-identical key helpers existed (models-list-result,
models-list-configured-static, harness/model-catalog). Export
resolveModelCatalogIdentityKey from the route-policy owner, collapse the
duplicate dedupe loops into dedupeByKey, make donor enrichment Map-based,
and inline the one-off harness-augment wrapper.
* fix(macos): restore custodian handoff for fresh activations
Landing every finish on the plain dashboard stranded the custodian
first-run flow (memory import, channels, permissions, hatch). Fresh
activations now hand off to custodian onboarding; live-verified
pre-existing setups reopen the normal dashboard, matching the removed
already-configured shortcut. Tests pin the destination per path.
Also isolate the post-startup Codex login test from developer machines:
ambient OPENAI_API_KEY and a real Codex login made it assert-fail.
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
Fix xAI barge-in truncation that reported 3760 ms after only 3700 ms of decoded audio.
Share exact audio-duration math across OpenAI, xAI, and Google while keeping provider-specific rounding at each caller.
* fix(voice-call): survive gateway in-process restart and stop CLI dead-ends
The gateway's in-process restart (SIGUSR1 config reload) reuses the cached
plugin registry, so service stop/start run on the same retained voice-call
registration. Generation fencing from #120289 treated that restart as a stale
actor: stop retired the generation forever, the next start silently bailed,
and every voicecall.* RPC answered UNAVAILABLE "runtime generation is
retired" while the webhook never rebound.
- Registrations now hold a replaceable generation: service start after stop
mints a fresh generation, takes over a running slot owned by a retired
predecessor, and reports start failures to service health instead of
silently returning.
- The voicecall CLI classifies gateway failures with typed guards instead of
message substrings: standalone/store fallback only when the gateway is
genuinely absent; reachable-but-failed (request errors, auth, timeout)
exits with actionable text; a standalone webhook port collision explains
that a running Gateway probably owns the port instead of raw EADDRINUSE.
- Plugin SDK gateway-runtime exports structural isGatewayTransportError /
isGatewayClientRequestError guards (+2 documented surface budget).
- Regression coverage: same-registration stop/start restart, retired-owner
takeover, typed CLI fallback classification, and a real token-auth gateway
server routing voicecall.status through callGatewayFromCli.
* refactor(voice-call): split CLI modules and dedupe gateway fallbacks
Collapse the four duplicated gateway-or-runtime command blocks (speak, dtmf,
end, continue fallback) into one generic runGatewayManagerCommand helper —
the continue command owns its legacy-method fallback and operation polling
via a gatewayCall closure, so the helper carries no per-command policy.
Smoke reuses the shared initiateVoiceCall path instead of a bespoke
fallback.
Split the 988-line cli.ts into concept modules (cli-gateway-call,
cli-call-log, cli-command-io) and drop its grandfathered max-lines
suppression plus the now-stale max-lines and assertion-safety baseline
entries (shrink-only ratchet maintenance).
Behavior-frozen: stdout/exit semantics unchanged; net -2 production LOC.
* fix(voice-call): redact gateway URLs in CLI operational errors
ClawSweeper P1: the operational-error formatter interpolated the raw
connectionDetails.url, so a configured gateway URL with userinfo or query
tokens would print credentials into terminal output. Redact the composed
message once with the canonical net-policy redactor (also covers
remote-controlled close-reason text), exported through the plugin SDK
gateway-runtime subpath (+1 documented surface budget). Regression test
covers a credential-bearing URL in both the URL and message fields.
* feat(agents): unify agent status into a durable progress_card
Replace the write-only update_plan to-do tool and the fragmented plan
rendering with one durable status artifact per session: progress_card
({plan?, markdown?}, replace-on-write, 8 KiB markdown / 50-step caps).
Cards persist in a lazy-additive session_progress_cards table in the
per-agent DB (no schema-version bump), broadcast progressCard.changed,
and render from the store with exactly one live placement per view
(session rail when visible, else the composer-adjacent bar); transcripts
collapse to one-line receipts, and the sidebar hovercard shows other
sessions' cards inline (markdown + <progress>, DOMPurify allowlist, no
iframes). The three stream-derived plan renderers and their dedup
heuristics are deleted.
Codex runs disable the native plan tool per thread
(tools.update_plan.enabled=false) and receive progress_card via the
dynamic-tool bridge; compaction restore now reinjects the card (steps +
bounded markdown). Card writes still emit the legacy plan stream event so
native apps and channels keep working until their per-platform
migrations. Policy names map update_plan -> progress_card; the shipped
tools.updatePlan=false kill switch is honored.
Net -277 production LOC; -480 test LOC.
* test(agents): regenerate Codex prompt snapshots for update_plan thread-config disable
* chore(protocol): allowlist progressCard.changed for native apps pending card migration
* fix(ci): repair progress card integration checks
* fix(codex): canonicalize native progress cards
* test(gateway): reconcile progress card method order
* test(codex): stabilize native approval fixture
Plugins holding a remote URL had no non-deprecated Plugin SDK path to turn it into managed media; saveMediaSource was only reachable through the deprecated media-runtime barrel while docs pointed at media-store. Re-export it from the focused subpath and move the bundled qa-channel plugin off the deprecated barrel. Maintainer decision: full saveMediaSource (local path + HTTP(S)) is the supported media-store contract.
Fixes#125259
* refactor(gateway,ui): one bounded display projection; delete marker strip sites
Persisted transcripts are marker-free since the write-boundary projection
(#124793), the historical migration (#124888), and TTS facts (#124913), so
display surfaces stop compensating. sessions.list.lastMessagePreview and its
siblings (sessions.preview/describe, TUI picker, sessions_list tool, MCP) now
share one bounded role-aware projection (240 chars, tool/system/thinking and
suppressed control replies excluded, directive-only rows fall through). The
web reply chip reads the typed openclawDelivery fact instead of parsing text;
chat.history preserves the field to the UI. Post-hoc display strips are
deleted across web/TUI/MCP/sessions-list; live streaming cleaners stay.
Stale gateway-protocol preview comments corrected; no schema change.
Assertion-safety baseline pruned for shrunk files (sanctioned direction).
Production net -173, tests net -137. Fixes the sidebar [[reply_to_current]]
preview leak and the empty-code-pill overstrip of quoted markers.
* fix(agents): preserve restart recovery transcript reads
* refactor(gateway): remove obsolete transcript exports
* fix(gateway): normalize injected delivery directives
* fix(ci): scope projection and recovery checks
* chore(ci): shrink plugin SDK surface budgets
* test: deflake loaded side question and worker checks
* test: align display projection CI fixtures
* style: format display projection fixture
* test(plugins): reset the whole plugin metadata lifecycle between tests
Tests that touch process-global plugin metadata were calling
clearCurrentPluginMetadataSnapshot, which drops the published snapshot but leaves
every memo registered through registerPluginMetadataProcessMemoLifecycleClear
populated. Those memos are not content-addressed: setup-registry keys its cached
registries on resolveCurrentSetupSnapshotCacheId, which returns the literal
"nosnap" whenever no snapshot is published. A registry cached by one file is
therefore served to any later file that also runs without a snapshot.
The affected lanes run isolate: false, so that is a live cross-file leak. It
matches the observed failure shape, e.g. setup-registry descriptor lookup failing
with "Cannot read properties of undefined (reading 'flatMap')" only inside a
shared worker and never in isolation.
Switch these files to clearPluginMetadataLifecycleCaches, which clears the
snapshot and every registered memo as one lifecycle unit. Two files kept both
calls; the narrow one is now redundant and removed.
current-plugin-metadata-snapshot.test.ts keeps the narrow primitive: it is the
unit test for that function.
No production change.
* test(plugins): drop the duplicate lifecycle reset in provider-runtime hooks
The blanket clearCurrentPluginMetadataSnapshot -> clearPluginMetadataLifecycleCaches
substitution collided with the lifecycle call this suite already made, so each
test boundary ran every registered metadata-cache clear twice. Keep one call per
hook. Addresses the ClawSweeper P3 finding on #125037.
* fix(security): harden canonical bounded JSON body reading
* refactor(admin): use canonical bounded JSON body reader
* refactor(telegram): use canonical bounded JSON body reader
The assertion-safety baseline prune for extensions/telegram/src/miniapp/routes.ts (2 to 1) is explicitly approved.
* test(security): cover canonical JSON body migrations
* refactor(plugin-sdk): name response-first body profile
* fix(telegram): flush miniapp body-limit responses before close
Split Signal approval routing into its concept-owned module and call the shared SDK binding helpers directly.
Reject persisted reaction targets when any allowed decision is invalid or duplicated, rather than retaining a valid subset from a corrupt transient record.
The Signal and iMessage channel plugins inferred approval prompts by
regex-sniffing rendered message text (header matching like 'Exec approval
required', /approve command parsing), violating the transport-only channel
doctrine: approval actions must stay typed until channel encoding.
The typed envelope already existed (channelData.execApproval with
approvalId/approvalKind/allowedDecisions) and every payload-level delivery
path consumed it; the regex paths were redundant re-derivation at the raw
send seam plus restart recovery for in-memory iMessage poll state.
- Signal: delete send-level prompt sniffing entirely; all approval sends
already flow through typed structured-payload or native-handler paths.
- iMessage: sendMessageIMessage takes a typed approvalPrompt binding
(id/kind/decisions) from the native approval handler instead of an
approvalKind flag plus text re-parsing.
- iMessage poller: persist pending poll targets in the plugin keyed store
so restart recovery no longer regex-scans chat history; typed recent-chat
discovery for handle-only DM targets stays. Split poll-target ownership
into approval-reaction-poll-targets.ts (max-lines).
- Plugin SDK: remove extractApprovalReactionPromptBinding — beta-only
surface, never in a stable release, so no deprecation window applies;
AGENTS.md now records that rule.
Accepted tradeoff: approval prompts delivered by a pre-upgrade process are
not rediscovered from chat text after restart (<=24h transient state;
persisted reaction bindings and event-driven tapbacks still work).
Peter Steinberger