* refactor(plugins): retire deactivate hook alias
* refactor(plugin-sdk): prune retired facade exports
* test(logging): isolate logger test controls
* refactor(logging): internalize file transport controls
* test(plugin-sdk): preserve retired facade coverage
* test(auto-reply): remove stale diagnostic imports
* refactor(logging): delete dead config-read guard
shouldSkipMutatingLoggingConfigRead had no production caller even on main;
it survived the dead-export scan only via logger's testApi re-export. The
test-isolation commit removed that mask, exposing the fossil. Delete the
guard, its test-only re-export, its mock entry, and its dedicated test file.
* refactor(plugin-sdk): retire due compatibility subpaths
* test(plugin-sdk): type group policy predicates
* refactor(plugin-sdk): split removed subpath records
* refactor(secrets): remove retired collector barrel
* test(plugin-sdk): tighten wildcard surface pin
* refactor(plugin-sdk): retire matrix facade metadata
* style(plugin-sdk): format facade metadata
* fix(ci): load channel setup contracts from source
Repair the main-owned regression from 99d662473c (Peter Steinberger): the new env-contract test could consume stale ignored dist metadata instead of the checked-in plugin declaration.
* test(plugin-sdk): refresh API baseline after rebase
Keep raw commands, paths, and provider errors out of ordinary chat while preserving explicit raw diagnostics and structured admin history.
Default command progress is status-only; `/verbose full` and `commandText: "raw"` retain diagnostic detail.
Co-authored-by: Ayaan Zaidi <hi@obviy.us>
Allow before_agent_reply plugins to declare host-enforced trigger eligibility so scheduled-only hooks do not block interrupted user-turn recovery. Keep omitted and malformed scopes fail-closed, scope both memory-core maintenance hooks, and cover three runner reload cycles through the public registration contract.
Refs: #111442
Source: #114836
Co-authored-by: Vincent Koc <vincentkoc@ieee.org>
* feat(memory): add provenance and recall metadata to the memory index
* feat(memory): provenance-gated promotion and capture hygiene
* feat(dreaming): LLM consolidation with deterministic gates, on by default
* feat(active-memory): deterministic recall lane with escalation default
* feat(memory): user model file and standing intents
* docs(memory): document the memory architecture
* fix(memory): live-QA fixes — metadata writers, provenance classes, intent scope, claim accumulation
* feat(sdk): always persist media facts and ship facts-first replacements for legacy Media* surfaces
PR 1 of the media legacy retirement program (audit-frozen, 4 PRs).
- Every media-bearing user turn now persists normalized __openclaw.media
facts unconditionally while continuing to emit the legacy top-level
Media* projection byte-identically (dual-write bridge; the conditional
shouldPersistStructuredMediaEntries gate now always includes media).
- New replacement APIs, shipped before any removal: typed hook media
facts (media[], originalMedia[], mediaStagingPending) on message
events; {{AttachmentPath}}/{{AttachmentUrl}}/{{AttachmentContentType}}/
{{AttachmentDir}}/{{AttachmentIndex}} template variables; focused
openclaw/plugin-sdk/media-local-roots subpath split out of the
deprecated agent-media-payload facade.
- Every legacy surface carries @deprecated naming its replacement, under
one named compatibility record media-legacy-projection with the
operator-approved removeAfter 2026-10-01 (two release trains; deletion
additionally gates on a clean published-plugin artifact sweep).
- Generic transcript append invariant documented; SDK migration, hooks,
and configuration docs updated to the facts-first path.
Writer golden matrix proves legacy bytes and model prompt bytes are
unchanged while nested facts become unconditional. 2,189 broad media
tests green; SDK api-baseline regenerated on fresh-env Testbox.
* feat(sdk): register media-local-roots subpath exports and deprecation metadata
Completes PR 1: package export map for openclaw/plugin-sdk/media-local-roots
plus the deprecated-subpath inventory and doc metadata entries for the
media-legacy-projection record.
* chore(sdk): track media-local-roots entrypoint and deprecated-export budgets
* fix(sdk): keep deprecated MSTeams buildMediaPayload re-export through the compat window
Deleting shipped runtime-api re-exports belongs to retirement PR 4 after
the media-legacy-projection window; PR 1 only deprecates. Also formats
the migration-guide schedule table.
* docs: regenerate docs map for media migration additions
* feat(cron): stream schedule sources (supervised command stdout)
Add gated argv stream schedules with bounded line batching and trigger.streamBatch composition.
Reuse the gateway ProcessSupervisor for source ownership, deterministic teardown, capped restart backoff, and schedule-key guarded batch execution. Expose additive protocol, CLI, tool, UI, docs, and generated snapshot surfaces without storage DDL.
Contract: stream schedules are event-driven, require cron.triggers.enabled, reject command payloads, and retain at most one bounded pending batch.
* fix(cron): reject retired stream source epochs at run admission
Thread an invalidatable per-owner source-generation token (ownerNonce.generation)
through cron.run admission alongside the schedule key. A batch handed to cron.run
under one source epoch can wait behind another run while its owner is stopped; a
disable→re-enable or A→B→A edit leaves the schedule key unchanged, so the key
check alone would admit the retired epoch's batch. The token is persisted in
job.state on every lifecycle write and compared at every admission site plus the
executeJobCore guard, so a stale epoch's batch is skipped.
Also fix direct stream-job mutations recording the wrong lifecycle status when
global cron is off but triggers are on: extract resolveStreamStopReason so the
direct path reports the remediable cron-disabled state like reconcile does.
* fix(cron): close stream admission windows from round-10 review
- Persist the retired source generation before draining stop teardown, so a
batch queued behind another cron run cannot gain admission during the up-to-10s
in-flight-batch wait (server-cron routed stop path).
- Add streamSourceGeneration to the closed gateway response schema (excluded from
the writable patch schema) so a running stream job passes strict result
validation without letting callers spoof source identity.
- Close the mutation-epoch ABA: track an eviction epoch so a snapshotted absent-0
is trusted as unchanged only when no LRU eviction happened during the await.
* refactor(cron): stream sources own a durable logical identity
Split the conflated restart-generation/admission token into two concepts:
a persisted streamSourceIdentity owned by cron store mutations (rotates on
enable/disable, source replacement, once-trigger auto-disable, and explicit
retirement; stable across supervised child restarts) and a watcher-local
process generation used only to fence stale child callbacks. Admission now
requires schedule key + identity together at every window, closing the
A-to-B-to-A and restart-flush races from review rounds 8-11.
Also: match-mode regexes now see raw source text (the [truncated] marker is
applied after matching), stop-timeout failures set the live restartExhausted
mirror so shutdown preserves the terminal diagnostic, and the watcher is
split into owner/output/registry modules under the max-lines budget.
* fix(cron): harden stream teardown and intake from round-4 review
An exit queued ahead of a requested stop no longer counts toward restart
exhaustion (the synchronous stop fence owns it), overlapping cron.stop and
stopAndDrain share one memoized shutdown drain instead of double-stopping
every owner, raw output intake is bounded at 4x the batch cap so normal
64 KiB pipe reads stop losing complete lines to OS chunk boundaries, and
persisted-shape quarantine coverage for unsafe match expressions is pinned.
* fix(cron): keep watcher-internal owner disposal from retiring live identity
Disposing an obsolete owner while a start replaces it is not a durable
removal; a retiring stop there rotated the live job's identity and stranded
the replacement behind the CAS ownership guard. Also align the docs with the
implemented match semantics: complete lines match on full text past the
batch cap, only intake-cut prefixes are unmatchable.
* fix(cron): make oversized-line matching independent of pipe chunking
Partial lines are retained up to the raw-intake bound rather than the
delivery cap, so a complete over-cap line matches identically whether it
arrives in one callback or several; only a line the intake bound itself cut
remains an unprovable prefix. Reconcile also contains schedule-replacement
stop failures per job, matching the other stop branches.
* fix(cron): bound assembled lines, drop stale payload override, barrier stopAll
- enforce the 4x raw-intake per-line cap while assembling split callbacks,
so an oversized line stays an unprovable prefix regardless of chunking
- stop passing the watcher-cached payload as a cron.run override; the run
snapshots the persisted payload under its admission lock
- stopAll waits for every owner stop to settle before surfacing failures
- split cron-stream-output interleaving tests into their own file (max-lines)
* fix(cron): address first full-CI round (lint, knip, schema test, unused param)
* chore(cron): refresh codex prompt snapshots for stream schedule schema
* fix(cron): keep the first clean line after an intake drop ending at a newline
* fix(cron): fence stream reconcile list snapshots against direct mutation routes
A cron.list snapshot captured across the reconcile await could be applied
after a direct add/update route already started the owner, stopping it as
removed and retiring its live identity. A mutation revision bumped at every
direct route start invalidates the stale snapshot; reconcile re-lists
(bounded) instead of applying it.
* style(cron): format stream owner imports
* fix(cron): discard severed stream prefixes at EOF
* fix(cron): retry failed stream shutdown drains
* fix(cron): honor stream stop fence after output drain
* chore(cron): refresh landing checks
* refactor(plugin-sdk): narrow wildcard barrels to explicit used exports
* refactor(tools): delete dead tool-planning module exposed by barrel narrowing
* fix(plugin-sdk): restore deprecation tag on OpenClawSchemaType alias
* test(agents): drop test for deleted runtime proxy module
* refactor(tools): trim descriptor types to cache consumers
* refactor(deadcode): harvest exports orphaned by barrel narrowing
* refactor(deadcode): harvest exports orphaned by barrel narrowing (rest)
* fix(agents): restore sdk imports and test markers via public predicate
* fix(plugin-sdk): named type re-exports in plugin-entry; trim types barrel precisely
* chore(plugin-sdk): account unmasked deprecated provider types in budgets
* fix(plugins): name star-only type rows for dts bundling
* fix(plugins): restore host-hook surface; unexport internal api compositions
* fix(plugins): named type imports for api composition; restore needed source exports
* fix(plugins): knip-visible type imports for registry surfaces
* test: adapt tests to privatized media and command internals
* fix(qa-lab): re-export snapshot conversation type
* style: format sessions sdk imports
* fix(plugins): restore smoke entry export; pin budgets to exact actuals
* fix(plugins): canonical smoke-entry import; drop orphaned root shims
* fix(plugins): allowlist manifest probe, repoint qa web import, drop dead browser barrels
* fix(plugin-sdk): pin codex auth marker and scaffold provider type
* fix(qa-lab): keep web-facing model-selection shim within boundary rules
* fix(plugin-sdk): preserve merged contracts through narrowed barrels
* chore(plugin-sdk): pin post-rebase surface budgets
Adds an `on-exit` cron schedule kind: a job fires once when a watched command/process
exits, via gateway ProcessSupervisor exit watchers. Covers CLI (`--on-exit`/`--on-exit-cwd`),
tool/protocol schema, RPC list-filter, Control UI + macOS read-only display, SQLite
round-trip, and origin-aware wake routing. Restart-safe one-shot (persists completion
before firing); platform-aware shell; bounded watched-command execution.
Squashed from 22 iterative commits for a clean rebase onto current main.
* fix(reply): deliver final reply when queued follow-up claims session; scope dedupe to routed thread
Two core bugs caused composed replies to be silently dropped (no delivery,
no error) when a second message arrived in the same thread mid-run:
1. dispatch-from-config: ensureDispatchReplyOperation only kept the
dispatch-owned operation authoritative while it had no result. Once
runReplyAgent completed the operation to drain queued follow-ups, a
second same-thread inbound could claim the session and the first final
reply would try to re-acquire the lane instead of finishing delivery,
deadlocking behind the queued work. Keep the dispatch-owned operation
authoritative through final delivery.
2. reply-payloads-dedupe: messaging-tool reply dedupe compared only the
channel target, not the routed thread, so a send in one thread could
suppress a later reply in a different thread. Thread the routed thread
id through buildReplyPayloads + follow-up delivery and only fall back to
channel-only matching for providers without a thread-aware suppression
matcher when neither side carries thread evidence.
Adds regression tests; existing Telegram topic-suppression behavior is
preserved by gating the thread guard to providers lacking a plugin matcher.
* fix(reply): preserve threaded message delivery evidence
* fix(reply): dedupe final payloads by delivery route
* fix(slack): preserve native send thread evidence
* fix(reply): preserve explicit reply thread evidence
* fix(reply): align explicit reply route dedupe
* fix(reply): preserve delivery lane through final dispatch
* fix(mattermost): preserve threaded tool send routes
* chore(plugin-sdk): refresh API baseline
* fix(reply): align final delivery route dedupe
* fix(reply): gate followups on final delivery
* fix(reply): keep send receipts private
* fix(reply): infer implicit message provider
* fix(reply): align routed threading policy
* fix(reply): preserve queued delivery context
* fix(reply): hydrate queued system event routes
* fix(reply): hydrate queued execution routes
* fix(reply): scope final delivery barriers
* fix(slack): preserve DM target aliases
* fix(reply): mirror resolved source thread routes
* fix(mattermost): retain delayed delivery barrier
* fix(codex): separate message routing from tool policy
* fix(reply): consume normalized Slack DM targets once
* fix(slack): remove stale target alias
* style(reply): satisfy changed lint gates
* fix(mattermost): preserve explicit reply targets
* test: align Slack reply branch checks
* fix(reply): persist overflow summaries to admitted session
---------
Co-authored-by: Peter Steinberger <steipete@gmail.com>
context.used_tokens / pct_used were derived from the snapshot's aggregate
prompt total (cacheRead+cacheWrite+input). Over a multi-call tool-loop turn
that is the run AGGREGATE, overstating window occupancy (often past 100%) so a
footer's context gauge pins full while /status shows the true ~7%.
Add two optional fields to PluginHookReplyUsageState and populate them in the
reply path:
- contextUsedTokens: the final call's prompt size (agentMeta.promptTokens) =
real end-of-turn occupancy, a point-in-time state, not the aggregate.
- lastUsage: the final model call's usage only (vs `usage`, the turn
aggregate), so a footer can render the last exchange's i/o + cache.
Both optional and additive; consumers fall back to the aggregate when absent
(correct for single-call turns). Renderer consumption lands separately (#89835).
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Addresses review feedback on #89629.
1) Provider-limit resolution no longer defaults to OAuth. getProviderUsageLimits
and getProviderUsageLimitsCached resolved with `credentialType ?? "oauth"`, and
the agent-runner snapshot call passed no credential type, so an api-key OpenAI
turn could borrow cached OAuth/ChatGPT usage windows. Drop the "oauth" default
(missing credential type => no OpenAI usage provider) and thread the turn's
authMode through at the call site. Adds provider-usage.limits.test.ts covering
api-key/no-credential (no fetch), oauth/token (resolves), and non-OpenAI.
2) usageState is documented as best-effort, present only on live dispatcher
delivery. Routed durable and recovered queue replays re-run this hook as a
stateless transform over the original payload (see QueuedDeliveryPayload); a
point-in-time usage snapshot is not stateless and would replay stale after a
restart, so it is intentionally omitted there. Consumers must treat the field
as optional.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Attach a per-turn execution snapshot to the reply_payload_sending hook as
`usageState`, so a plugin (or the future in-core /usage renderer) can render a
per-response usage readout as a pure consumer of the contract — no side calls.
Recorded in agent-runner, consumed in dispatch. Fields: provider, model,
resolvedRef/requested, reasoningEffort, fastMode, fallbackUsed, is_override
(overrideSource), authMode, compactionCount, contextTokenBudget, token usage,
turn cost (USD), duration, owning agentId/sessionId, chatType, the agent
identity (name/emoji), and the active provider's subscription `limits` windows.
reply_payload_sending is the one reply hook universal across every surface
(incl. the Codex app-server, which emits no llm_output/agent_end), so it is the
correct harness-agnostic place for per-turn usage. Limits are resolved by a
core-internal non-blocking SWR helper (src/infra/provider-usage.limits.ts) and
attached to the snapshot — no new plugin-SDK accessor. All fields optional/additive.
Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>
Fixes#88521.
Expose finalized inbound reply metadata on plugin-visible hook payloads so before_dispatch and message hooks can implement reply-aware behavior without channel-specific workarounds.
Peter Steinberger