…aw#195)

…aw#197)

Add useGatewayEnabled() and useGatewayModeKey() shared hooks that centralize
the gateway mode logic:
- Production: always returns true (gateway is the primary data source)
- Development: returns true only when user has toggled 'Live Gateway' on

Migrated all 9 files that had inline import.meta.env.DEV && useLiveGateway
checks to use the shared hook instead:
- useAgentMutations.ts (4 mutation hooks)
- useConversationMutations.ts (useLiveMode helper)
- useAgentStatus.ts (dashboard hook)
- useConversations.ts (4 query hooks)
- useAgents.ts (3 query hooks)
- useNodes.ts (3 query hooks)
- __root.tsx (gateway auth guard)
- main.tsx (gateway provider)

This makes production web builds fully functional by inverting the dev-only
gating that previously made the gateway always disabled in production.

…invalid-skills-rpcs

chore: remove invalid skill RPCs, add install/uninstall metadata to all skills

…startup

fix: schema migrations not being discovered properly

…penclaw#201)

- Create RouteErrorFallback component for TanStack Router errorComponent
  - Shows which page crashed with derived route name
  - Try Again button to reset the route error
  - Go Back button to navigate back
  - Collapsible error details (auto-shown in power user mode)
  - Copy error details to clipboard
- Wire errorComponent to all 17 major routes
  - Isolates errors per-page so one broken view doesn't take down the app
  - Uses TanStack Router's built-in error boundary mechanism
- Export RouteErrorFallback + ErrorState from composed components index

…mains (openclaw#200)

Implements cross-domain search hook for the web UI:
- Fuzzy matching with scoring: exact > starts-with > substring > word-prefix > char-sequence
- Searches across agents, sessions, goals, decisions, cron jobs, and navigation
- Category grouping with icons and ordered display
- Debounced input with configurable delay
- Results sorted by relevance score with per-category and total limits
- 15 unit tests for fuzzy matching and category metadata
- Exported from hooks/index.ts for easy consumption

Implement a comprehensive scoring system that evaluates whether experiences
should become long-term memories based on five weighted factors:

- **Novelty**: Is this new information or repetition?
- **Impact**: Does this change understanding, behavior, or system state?
- **Relational**: Does this connect to known entities/people/projects?
- **Temporal**: Is this time-sensitive or evergreen?
- **User Intent**: Was this explicitly marked as important?

Features:
- Configurable weights for each factor (normalized to sum to 1.0)
- Named threshold profiles: balanced, aggressive, conservative, minimal
- Override rules for specific tools (glob-pattern matching)
- evaluateMemoryRelevance() as primary entry point
- shouldCapture(), shouldPersistToGraph(), shouldUseLlmEval() helpers
- Backward-compatible evaluateHeuristic() now delegates to scoring system
- New evaluateHeuristicDetailed() returns full scoring breakdown
- 37 comprehensive tests covering all factors and edge cases

Files:
- extensions/meridia/src/meridia/scoring/types.ts
- extensions/meridia/src/meridia/scoring/config.ts
- extensions/meridia/src/meridia/scoring/factors.ts
- extensions/meridia/src/meridia/scoring/index.ts
- extensions/meridia/src/meridia/scoring/scoring.test.ts
- extensions/meridia/src/meridia/evaluate.ts (updated)

…7) (openclaw#202)

Implements a comprehensive, configurable scoring system for determining
which experiences should become long-term memories.

## New Module: scoring/

- **types.ts**: Full type definitions for factors, breakdowns, configs
- **factors.ts**: Five factor scorers (novelty, impact, relational, temporal, userIntent)
- **scorer.ts**: Composite scorer with tool/pattern overrides, threshold profiles
- **defaults.ts**: Default weights, profiles (standard/aggressive/conservative), overrides
- **index.ts**: Barrel exports

## Factor Scoring Model

- **Novelty (0.25)**: Detects repetitive vs. new tool usage patterns
- **Impact (0.30)**: Classifies tool types (high/medium/low impact), error detection, meta keywords
- **Relational (0.15)**: Entity detection via tags, keywords, file paths, URLs
- **Temporal (0.10)**: Time-of-day awareness, burst detection, pacing analysis
- **User Intent (0.20)**: Explicit marks, manual captures, intent keywords

## Override System

- Tool-specific rules: fixedScore, minScore, maxScore, weightMultiplier
- Pattern-based rules: error boost, large result detection, tag matching
- User intent override: highest priority

## Integration

- Updated evaluate.ts with evaluateRelevance() bridge function
- Updated config.ts with scoring config resolution (backwards compatible)
- Re-exports for convenience: shouldCaptureMultiFactor, isHighValueMultiFactor, etc.
- breakdownToTrace() for compact trace event embedding

## Tests

- 48 tests across factors.test.ts and scorer.test.ts
- All passing

…te w/cron

feat(meridia): multi-factor memory relevance scoring system

feat: add configurable subsystem debug log suppression

* refactor(tools): create shared tool execution module

* refactor(pi): use shared tool execution in pi-tool-definition-adapter

* refactor(sdk): use shared tool execution in tool-bridge

* test(tools): add tests for shared tool execution module

* docs: add unified tool execution coverage gap analysis

* fix(auto-reply): use unified tool execution for inline skill commands

* docs(tools): document wrapper reliance on runtime adapters

* test(auto-reply): add inline skill command error handling tests

* docs: update gap analysis with completed fixes

* fix: missing stuff

)

- Consolidate /agents/dashboard and /agent-status into a single route with grid/list layout toggle
- Reduce summary cards from 7-8 to 4 primary + conditional stalled/errored alert chips
- Declutter agent cards: combined session counts, cron badge in header, no redundant health metric
- Fix detail panel on mobile: use Sheet overlay on sub-lg screens instead of hidden panel
- Lighten activity timeline: borderless rows for completed items, full Card only for live tasks
- Add inline metadata stats (duration, tokens, cost) on timeline rows
- Batch consecutive search/code activities into collapsible groups
- Remove hardcoded mock data fallback in AgentActivityTab; show empty state instead
- Add keyboard navigation (ArrowUp/Down, Escape) and aria-live for screen readers
- Group developer actions (Copy ID, Copy JSON) into overflow dropdown menu
- Promote session section above generic actions in detail panel
- Extract shared format utilities (formatTokenCount, formatRelativeTime, etc.) into lib/format.ts
- Improve Lit dashboard: CSS grid metrics, channel badges, right-aligned tokens, hover highlights

Co-authored-by: Claude <[email protected]>

Evaluates the original Meridia V2 architecture proposal against the
actual codebase state and existing architecture docs (ARCH.md,
COMPONENT-MAP.md, 7 component deep-dives). Produces a refined plan
that builds on the designed 14-component system rather than
reinventing it.

Key refinements over the original proposal:
- Two-pass evaluation (cheap gate, expensive phenomenology only on
  captured events) instead of full LLM extraction on every tool result
- Compaction-based graph sync with session-end mini-compaction instead
  of per-event Graphiti writes
- Vector search adapter interface (Graphiti vector first, pgvector
  optional) instead of hard PostgreSQL dependency
- Includes missing components (normalizer, gates, sanitization) the
  original proposal omitted
- Migrates to ExperienceKit type from COMPONENT-MAP rather than
  extending MeridiaExperienceRecord
- 5-phase implementation plan with dependency ordering

https://claude.ai/code/session_01Ko4xrCYytJBNW5VAxzfemE

…l, and reconstitution

Implements the Meridia V2 plan across all 14 components from COMPONENT-MAP.md:

**New modules (26 files):**
- event/normalizer: Component 1 — typed MeridiaEvent envelopes from hook payloads
- gates/budget: Component 2 — rate limits and capture budgets (extracted from hook)
- phenomenology/{taxonomy,prompt,heuristic,extractor}: Component 4 — two-pass
  phenomenological extraction with controlled vocabularies (17 emotions, 5 engagement
  qualities, 12 texture metaphors, 5 sensory channels from experiential-engine docs)
- artifacts/collector: Component 5 — file/link/media reference extraction
- kit/{builder,resolver}: Component 6 + Phase 5 — ExperienceKit assembly and
  meridia://<id> URI resolution
- sanitize/redact: Component 12 — secret redaction and payload truncation
- retrieve/{intent,ranker,hybrid,vector-adapter}: Component 10 — hybrid retrieval
  with blended ranking (significance, recency, diversity, source score)
- reconstitution/{engine,pack-builder}: Component 11 — structured ReconstitutionPack
  replacing bullet-list format with narrative, anchors, uncertainties, approach guidance
- fanout/dispatcher: Component 9 — fire-and-forget async dispatch with error isolation

**Updated existing files:**
- types.ts: Added Phenomenology, CaptureDecision, ArtifactRef, ExperienceKit,
  ReconstitutionPack types; added phenomenology field to legacy record
- hooks/experiential-capture: Uses normalizer, gates, phenomenology extractor,
  artifact collector, kit builder, and sanitizer
- hooks/meridia-reconstitution: Uses enhanced reconstitution engine with legacy fallback
- db/backends/sqlite.ts: Schema v2 migration (phenomenology columns), updated FTS
  searchable text, v1→v2 migration is idempotent
- meridia-search-adapter: Implements readFile() for meridia:// URI resolution,
  includes phenomenology in search snippets
- tools: Fixed pre-existing async/await issues across all three experience tools
- tests: Fixed async/await in sqlite and reflect tool tests, updated schema assertions

https://claude.ai/code/session_01Ko4xrCYytJBNW5VAxzfemE

…#208)

* feat: added work-queue workflow workers and a worker manager, integrate w/cron

* fix: address 12 logic flaws in workflow engine, adapter, phases, and cron types

- engine: capture phase before clobbering in catch handler (#1)
- engine: remove dead `state.plan ?? plan` fallback (#2)
- engine: mark workflow failed when all execution nodes fail (#6)
- adapter: fix retry count off-by-one (attemptNumber vs retryCount) (#3)
- adapter: clean up abort listener in sleep on timeout (#10)
- discover: fix batch/entries index mismatch when spawns fail (#4)
- execute: add cycle detection to topological sort (#5)
- review: add autoApproved flag to distinguish fallback approvals (#7)
- plan: add comment clarifying sessionKey reuse across repair attempts (#8)
- decompose: remove unused model/maxPhases/maxTasksPerPhase/maxSubtasksPerTask opts (#9)
- types: add autoApproved to ReviewIteration (#7)
- cron/state: use discriminated union for CronEvent (CronJobEvent | CronHealthEvent) (#12)
- tests: add WorkflowWorkerAdapter test suite (8 tests) (#11)

https://claude.ai/code/session_01L8kquwpmUh5zmU9S4MHgPu

---------

Co-authored-by: Claude <[email protected]>

…aSearchAdapter

Fixes P1 and P2 issues from code review:
- search() now properly awaits backend.searchRecords() (was calling .map on a Promise)
- status() now async with await on backend.getStats() (was reading undefined properties)

Resolved conflicts by accepting origin/main versions which include:
- Type-safe discriminated unions for CronEvent (job vs health events)
- Better workflow execution semantics (attemptNumber vs retryCount)
- Improved error handling and phase tracking in workflow engine
- Bug fixes for discovery question-to-result alignment
- Cycle detection in topological sort
- Auto-approval tracking in review iterations

All conflicts resolved in favor of main's improvements.

Co-Authored-By: Claude Sonnet 4.5 <[email protected]>

…ntial backoff

- RetryableError: composable error boundary with visual retry feedback
- useRetry: React hook with exponential backoff, jitter, countdown timer
- withRetry: standalone retry utility for non-React contexts
- Comprehensive test suite with 10 test cases
- Card and inline variants, auto-retry option, cancel/reset controls

Closes P4 Error Handling UX: retry patterns for failed operations

- GlobalSearchDialog: full-featured search overlay powered by useGlobalSearch hook
- Category filter chips for narrowing results by type (agents, sessions, goals, etc.)
- Recent searches persisted to localStorage
- Keyboard shortcut Cmd+Shift+F / Ctrl+Shift+F to open
- Search button added to sidebar with keyboard hint
- Score-based ranking with grouped category results
- Comprehensive test suite for keyboard shortcut hook

Wires up the existing useGlobalSearch hook into a production-ready UI.
Addresses P4 Search & Filtering from BACKLOG.md.

…and Obsidian probes (openclaw#382)

Adds a composable dependency health monitoring system:

- types.ts: DependencyHealthStatus, DependencyHealthProbe interfaces
  with tier (core/channel/integration/infrastructure) and probe mode (active/passive)
- registry.ts: Global probe registry with register/unregister/snapshot/refresh APIs
- probes/graphiti.ts: Active probe with periodic polling via GraphitiClient.health()
- probes/gmail.ts: Passive probe reading Gmail watcher state
- probes/channels.ts: Dynamic probe creation per enabled channel plugin
- probes/obsidian.ts: Passive probe reading Obsidian vault runtime state
- index.ts: Barrel export for all types, registry functions, and probe factories
- __tests__/registry.test.ts: 9 unit tests covering full registry lifecycle

Design: Probes are lazy-registered, support both sync cached reads
(getStatus) and async forced refresh. Active probes can start/stop
periodic polling. The registry is a singleton for gateway-wide access.

)

- server-startup.ts: Call initDependencyHealthProbes(cfg) after Obsidian init
- server-close.ts: Call stopDependencyHealthProbes() on shutdown
- commands/health.ts: Add dependencies[] to HealthSummary, populate from
  probe registry (refresh on active probe, cached read otherwise)

Completes the integration started in PR openclaw#382.

…law#387)

- Add preserveRunningForJobIds option to recomputeNextRuns() to prevent
  stuck-run cleanup from clearing running markers on jobs that are still
  claimed and waiting to execute in the same batch
- Fix heartbeat status check: 'error' → 'failed' to match actual API
- Add comprehensive regression test suite (10 tests) covering:
  - Schedule change recalculation
  - Force mode execution
  - Isolated job wake time scheduling
  - allowUnsafeExternalContent persistence on payload patches
  - Far-future schedule timer delay capping
  - Hot-loop prevention while run in progress
  - Manual run skip while timer-triggered run active
  - wakeMode now → queued heartbeat degradation when busy
  - Per-job start time and duration for batched jobs
  - Claimed running markers not cleared mid-batch after long jobs

openclaw#390)

* feat: CRN module — ClawdBrain Resource Names with ref kinds and canonicalization

Implements the CRN (ClawdBrain Resource Name) system:

- crn:{version}:{service}:{scope}:{resource-type}:{resource-id}
- Greedy resource-id field (6th colon onward) for embedded colons
- Service registry with URL patterns for codex-web, claude-web, github, slack, notion
- Bidirectional URL ↔ CRN conversion with round-trip fidelity
- EntityRef canonicalization: auto-enriches refs with CRN and resolved URI
- Extended RefKind enum with external platform kinds:
  codex-web:task, claude-web:task, claude-web:project, github:pr, github:issue,
  github:repo, github:commit, slack:channel, slack:message, notion:page, notion:database

84 tests covering parse, registry, canonicalization, and round-trips.

* chore: fixes

Add a new "Debug & Logging" section in the Admin/Power User settings
that provides a full UI for tuning openclaw.json logging and debugging
configuration. Gated behind Power User Mode.

Covers:
- Log level / console level / console style selectors
- Redact sensitive toggle
- Enhanced logging feature toggles (tool errors, perf outliers, etc.)
- Performance threshold sliders (tool call, agent turn, gateway, db op)
- Token warning threshold sliders
- Tool journal config (enable, file path, retention, redaction)
- Suppress subsystem debug logs tag list
- Channel debug overrides (add/remove with verbose/debug/trace/suppress)
- Feature debug overrides (add/remove with verbose/debug/trace/suppress)
- Sticky save/reset bar with dirty state tracking

https://claude.ai/code/session_011WdpEacHmK1fMQMFiD3Szu

Co-authored-by: Claude <[email protected]>

* fix: update Claude task URL format to /code/ path

Claude web task URLs use the format https://claude.ai/code/session_{id}
rather than https://claude.ai/chat/{id}. Updated the base URL in the
canonicalizer, builder, and tests accordingly.

* more work queue/work-item changes

* CRN: preserve case for external task IDs

…nclaw#391)

* Web: enhance Notion connection UX

* Web: harden Notion token test

* feat: Graphiti health probe + search-manager test fix

- Add periodic Graphiti health probe (configurable interval, 3-failure threshold)
- Gate entity-extract and graph writes behind health probe check
- Wire probe start/stop into gateway lifecycle
- Add healthProbeIntervalMinutes config option to MemoryGraphitiConfig
- Fix pre-existing test failure in search-manager.test.ts (ComposableMemoryManager
  creates new instances each call, so toBe reference check was incorrect)
- Add 10 comprehensive tests for health-probe.ts

* fix: minor pr comment

…claw#386)

* Web: add Obsidian connection setup

* Web: align Obsidian REST API defaults

* fix: improve debug logging for Graphiti client and startup service URLs

- Add URL to Graphiti client error logs for ingest and search operations
- Log configured external service endpoints (Graphiti, Obsidian, QMD) at startup
- Helps debug connectivity issues by showing which URLs are being targeted

* feat: add Goal Progress Panel to Overseer view

New component that provides detailed per-phase progress tracking for goals:
- Overall goal completion percentage with animated progress bar
- Status distribution visualization (done/in-progress/blocked/pending)
- Per-phase collapsible progress bars with stacked status segments
- Individual task listing with dependency indicators
- Velocity metrics (tasks/day) and estimated time remaining

Files added:
- ui/src/ui/views/overseer-progress.ts — Goal progress panel component
- ui/src/styles/overseer.css — Extracted overseer styles + new progress panel CSS

Integration:
- Added expandedPhaseIds state + toggleProgressPhase to overseer controller
- Wired renderGoalProgress into main overseer view (shows when goal has plan)
- Added CSS import to main styles.css

* feat: work queue logging enhancemenets

* chore: unused

…hod catalog from handlers (openclaw#392)

* Gateway: align overseer methods

* Gateway: include overseer goal startedAt

…w#397)

The finalizeJobRun function expects startedAtMs in its opts parameter,
but ops.ts was passing runAtMs which doesn't exist in the type. This
caused the plugin-sdk:dts build step to fail with TS2353.

…w#398)

The finalizeJobRun function expects startedAtMs in its opts parameter,
but ops.ts was passing runAtMs which doesn't exist in the type. This
caused the plugin-sdk:dts build step to fail with TS2353.

…stions, threshold validation, and power-user gating (openclaw#396)

* feat: add Debug & Logging config section to Settings UI

Add a new "Debug & Logging" section in the Admin/Power User settings
that provides a full UI for tuning openclaw.json logging and debugging
configuration. Gated behind Power User Mode.

Covers:
- Log level / console level / console style selectors
- Redact sensitive toggle
- Enhanced logging feature toggles (tool errors, perf outliers, etc.)
- Performance threshold sliders (tool call, agent turn, gateway, db op)
- Token warning threshold sliders
- Tool journal config (enable, file path, retention, redaction)
- Suppress subsystem debug logs tag list
- Channel debug overrides (add/remove with verbose/debug/trace/suppress)
- Feature debug overrides (add/remove with verbose/debug/trace/suppress)
- Sticky save/reset bar with dirty state tracking

https://claude.ai/code/session_011WdpEacHmK1fMQMFiD3Szu

* fix: address data loss, UX gaps, and missing fields in debug/logging config

Fixes:
- Preserve ALL config fields on save (file, redactPatterns, journal.toolFilter,
  enhanced.gatewayHealthSuppressMethods) via index signature and full-object
  patching — previously saving clobbered unmodeled fields
- Fix dead ternary in enhanced logging defaults (both branches were `true`)
- Add proper error state when gateway/config fails to load
- Add beforeunload guard to prevent losing unsaved changes on navigation
- Token threshold cross-validation: prevent warning >= critical with inline
  error message and disabled save button
- Gate "Debug & Logging" nav item behind power user mode in both desktop
  and mobile navs (was visible to all users before)

New config fields exposed:
- logging.file (log file path)
- logging.redactPatterns (custom regex redaction patterns)
- logging.journal.toolFilter (scope journal to specific tools)
- logging.enhanced.gatewayHealthSuppressMethods (suppress health logs
  for specific RPC methods)

UX improvements:
- Channel autocomplete suggestions from useChannelsStatusFast() gateway data
- Feature autocomplete suggestions from known codebase feature IDs
- Extracted reusable TagListEditor component for string[] config fields
- Quick-add badge chips for one-click channel/feature addition

https://claude.ai/code/session_011WdpEacHmK1fMQMFiD3Szu

* tests: relax cron timer timezone expectation

* Fix journal toggle default and tool filter help text in DebugLoggingSection (openclaw#399)

- Default the journal enabled toggle to true (was false), matching the
  runtime behavior where startJournalSubscriber treats undefined as enabled
- Update tool filter help text to reflect actual backend default
  (DEFAULT_JOURNAL_TOOLS subset) instead of claiming "all tools"
- Expand journal sub-settings when enabled field is unset (consistent
  with the new default-true toggle)

https://claude.ai/code/session_011jGYfDDmG4hSUMkct6YsjR

Co-authored-by: Claude <[email protected]>

---------

Co-authored-by: Claude <[email protected]>

- TelegramRichMessage tool: renders semantic patterns to Telegram HTML + InlineKeyboardMarkup
- Supports all 7 patterns: multiple_choice, confirmation, task_proposal, action_items, status, progress, info_grid
- TelegramResponseStore: tracks pending interactive requests with timeout handling
- 11 tests passing for rich message tool
- Response store for shared use by all interactive tools

…eractive tools

- AskTelegramConfirmation: Confirm/Cancel inline buttons, blocks until response
  - Custom labels, primary/danger style, message update after response
  - 9 tests passing
- AskTelegramQuestion: Single-choice and multi-choice inline keyboard
  - Multi-choice with toggle mode and Done button
  - Message update to show selected answer
  - 8 tests passing
- Both tools use callback router + response store integration