@swarmclawai/swarmclaw 1.9.6 → 1.9.7

package/README.md

@@ -399,6 +399,16 @@ Operational docs: https://swarmclaw.ai/docs/observability
 
 ## Releases
 
+### v1.9.7 Highlights
+
+Bundled eval-gate release: approved baselines, regression checks, and Quality Center release gates for repeatable eval evidence.
+
+- **Eval regression baselines.** Operators can snapshot the latest scenario or suite score as an approved baseline with minimum score and regression allowance settings.
+- **Release gate API.** `/api/eval/gate` compares current eval evidence against thresholds and baselines, while `/api/eval/baselines` lists and updates approved baselines.
+- **CLI gate checks.** `swarmclaw eval gate`, `swarmclaw eval baselines`, and `swarmclaw eval baseline-set` expose the same release-gate workflow from automation.
+- **Quality Center gate panel.** Eval Lab now shows pass/warn/fail status, latest-run coverage, current score, baseline score, regression points, and actionable checks.
+- **Public-source hygiene.** Generic implementation comments now describe SwarmClaw behavior without naming internal comparison sources.
+
 ### v1.9.6 Highlights
 
 Bundled eval-environment release: validation preflights, deterministic eval workspaces, and clearer operator readiness before spending run budget.

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@swarmclawai/swarmclaw",
-  "version": "1.9.6",
+  "version": "1.9.7",
   "description": "Build and run autonomous AI agents with OpenClaw, Hermes, multiple model providers, orchestration, delegation, memory, skills, schedules, and chat connectors.",
   "main": "electron-dist/main.js",
   "license": "MIT",
@@ -87,7 +87,7 @@
     "test:cli": "node --test src/cli/*.test.js bin/*.test.js scripts/electron-after-pack.test.mjs scripts/ensure-sandbox-browser-image.test.mjs scripts/postinstall.test.mjs scripts/run-next-build.test.mjs scripts/run-next-typegen.test.mjs",
     "test:setup": "tsx --test src/app/api/setup/check-provider/route.test.ts src/lib/server/provider-model-discovery.test.ts src/components/auth/setup-wizard/utils.test.ts src/components/auth/setup-wizard/types.test.ts src/hooks/setup-done-detection.test.ts src/lib/setup-defaults.test.ts src/lib/server/storage-auth.test.ts src/lib/server/storage-auth-docker.test.ts",
     "test:openclaw": "tsx --test src/lib/openclaw/openclaw-agent-id.test.ts src/lib/openclaw/openclaw-endpoint.test.ts src/lib/server/agents/agent-runtime-config.test.ts src/lib/server/build-llm.test.ts src/lib/server/connectors/connector-routing.test.ts src/lib/server/connectors/openclaw.test.ts src/lib/server/connectors/swarmdock.test.ts src/lib/server/gateway/protocol.test.ts src/lib/server/gateways/gateway-topology.test.ts src/lib/server/llm-response-cache.test.ts src/lib/server/mcp-conformance.test.ts src/lib/server/openclaw/agent-resolver.test.ts src/lib/server/openclaw/deploy.test.ts src/lib/server/openclaw/skills-normalize.test.ts src/lib/server/session-tools/openclaw-nodes.test.ts src/lib/server/session-tools/swarmdock.test.ts src/lib/server/tasks/task-quality-gate.test.ts src/lib/server/tasks/task-validation.test.ts src/lib/server/tool-capability-policy.test.ts src/lib/providers/openai.test.ts src/lib/providers/openclaw-exports.test.ts src/app/api/gateways/topology-route.test.ts src/app/api/openclaw/dashboard-url/route.test.ts",
-    "test:runtime": "tsx --test src/lib/a2a/agent-card.test.ts src/lib/strip-internal-metadata.test.ts src/lib/provider-sets.test.ts src/lib/providers/opencode-cli.test.ts src/lib/providers/cli-provider-metadata.test.ts src/lib/providers/cli-utils.test.ts src/lib/providers/generic-cli.test.ts src/lib/server/agents/delegation-advisory.test.ts src/lib/server/cli-provider-readiness.test.ts src/lib/server/provider-health.test.ts src/lib/server/mcp-gateway-runtime.test.ts src/lib/server/mcp-connection-pool.test.ts src/lib/server/knowledge-sources.test.ts src/lib/server/extension-managed-resources.test.ts src/lib/server/eval/environment-plan.test.ts src/lib/server/chat-execution/chat-execution-grounding.test.ts src/lib/server/chat-execution/chat-turn-preparation.test.ts src/lib/server/chat-execution/iteration-timers.test.ts src/lib/server/chat-execution/post-stream-finalization.test.ts src/lib/server/chat-execution/reasoning-tag-scrubber.test.ts src/lib/server/chats/clear-undo-snapshots.test.ts src/lib/server/connectors/email.test.ts src/lib/server/protocols/protocol-service.test.ts src/lib/server/runtime/run-ledger.test.ts src/lib/server/runtime/queue-retry-policy.test.ts src/lib/server/runs/run-brief.test.ts src/lib/server/operations/operation-pulse.test.ts src/lib/server/artifacts/artifact-resolver.test.ts src/lib/server/observability/otel-config.test.ts src/lib/server/safe-parse-body.test.ts src/lib/server/missions/mission-templates.test.ts src/lib/server/sharing/share-link-repository.test.ts src/lib/server/sharing/share-resolver.test.ts src/lib/server/tasks/task-execution-workspace.test.ts src/lib/server/tasks/task-service.test.ts src/lib/server/session-tools/execute.test.ts src/lib/server/session-tools/manage-tasks.test.ts src/lib/app/view-constants.test.ts src/lib/quality/quality-summary.test.ts src/app/api/approvals/route.test.ts src/app/api/agents/agents-route.test.ts src/app/api/tasks/tasks-route.test.ts src/app/api/tasks/task-workspace-route.test.ts src/app/api/chats/chat-route.test.ts src/app/api/chats/clear-route.test.ts src/app/api/chats/compact-route.test.ts src/app/api/chats/context-status-route.test.ts src/app/api/connectors/connector-doctor-route.test.ts src/app/api/extensions/managed-resources/route.test.ts src/app/api/healthz/route.test.ts src/app/api/logs/route.test.ts src/app/api/portability/export/route.test.ts src/app/api/portability/import/route.test.ts src/app/api/providers/[id]/route.test.ts src/app/api/tts/route.test.ts",
+    "test:runtime": "tsx --test src/lib/a2a/agent-card.test.ts src/lib/strip-internal-metadata.test.ts src/lib/provider-sets.test.ts src/lib/providers/opencode-cli.test.ts src/lib/providers/cli-provider-metadata.test.ts src/lib/providers/cli-utils.test.ts src/lib/providers/generic-cli.test.ts src/lib/server/agents/delegation-advisory.test.ts src/lib/server/cli-provider-readiness.test.ts src/lib/server/provider-health.test.ts src/lib/server/mcp-gateway-runtime.test.ts src/lib/server/mcp-connection-pool.test.ts src/lib/server/knowledge-sources.test.ts src/lib/server/extension-managed-resources.test.ts src/lib/server/eval/baseline.test.ts src/lib/server/eval/environment-plan.test.ts src/lib/server/chat-execution/chat-execution-grounding.test.ts src/lib/server/chat-execution/chat-turn-preparation.test.ts src/lib/server/chat-execution/iteration-timers.test.ts src/lib/server/chat-execution/post-stream-finalization.test.ts src/lib/server/chat-execution/reasoning-tag-scrubber.test.ts src/lib/server/chats/clear-undo-snapshots.test.ts src/lib/server/connectors/email.test.ts src/lib/server/protocols/protocol-service.test.ts src/lib/server/runtime/run-ledger.test.ts src/lib/server/runtime/queue-retry-policy.test.ts src/lib/server/runs/run-brief.test.ts src/lib/server/operations/operation-pulse.test.ts src/lib/server/artifacts/artifact-resolver.test.ts src/lib/server/observability/otel-config.test.ts src/lib/server/safe-parse-body.test.ts src/lib/server/missions/mission-templates.test.ts src/lib/server/sharing/share-link-repository.test.ts src/lib/server/sharing/share-resolver.test.ts src/lib/server/tasks/task-execution-workspace.test.ts src/lib/server/tasks/task-service.test.ts src/lib/server/session-tools/execute.test.ts src/lib/server/session-tools/manage-tasks.test.ts src/lib/app/view-constants.test.ts src/lib/quality/quality-summary.test.ts src/app/api/approvals/route.test.ts src/app/api/agents/agents-route.test.ts src/app/api/tasks/tasks-route.test.ts src/app/api/tasks/task-workspace-route.test.ts src/app/api/chats/chat-route.test.ts src/app/api/chats/clear-route.test.ts src/app/api/chats/compact-route.test.ts src/app/api/chats/context-status-route.test.ts src/app/api/connectors/connector-doctor-route.test.ts src/app/api/extensions/managed-resources/route.test.ts src/app/api/healthz/route.test.ts src/app/api/logs/route.test.ts src/app/api/portability/export/route.test.ts src/app/api/portability/import/route.test.ts src/app/api/providers/[id]/route.test.ts src/app/api/tts/route.test.ts",
     "test:builder": "tsx --test src/features/protocols/builder/utils/nodes-to-template.test.ts src/features/protocols/builder/utils/template-to-nodes.test.ts src/features/protocols/builder/validators/dag-validator.test.ts",
     "test:e2e": "node --import tsx scripts/browser-e2e-smoke.ts",
     "test:mcp:conformance": "node --import tsx ./scripts/mcp-conformance-check.ts",

package/src/app/api/eval/baselines/route.ts

@@ -0,0 +1,55 @@
+import { NextResponse } from 'next/server'
+import { z } from 'zod'
+import { evaluateEvalGate, listEvalBaselinesForAgent, setEvalBaseline } from '@/lib/server/eval/baseline'
+import { errorMessage } from '@/lib/shared-utils'
+
+const BaselineSchema = z.object({
+  agentId: z.string().min(1),
+  scenarioId: z.string().min(1).nullable().optional(),
+  suite: z.string().min(1).nullable().optional(),
+  minPercent: z.number().min(0).max(100).nullable().optional(),
+  maxRegressionPoints: z.number().min(0).max(100).nullable().optional(),
+  label: z.string().max(160).nullable().optional(),
+  notes: z.string().max(1_000).nullable().optional(),
+})
+
+export async function GET(req: Request) {
+  try {
+    const { searchParams } = new URL(req.url)
+    const agentId = searchParams.get('agentId')
+    return NextResponse.json(listEvalBaselinesForAgent(agentId))
+  } catch (err: unknown) {
+    return NextResponse.json(
+      { error: errorMessage(err) },
+      { status: 500 },
+    )
+  }
+}
+
+export async function POST(req: Request) {
+  try {
+    const body: unknown = await req.json()
+    const parsed = BaselineSchema.safeParse(body)
+    if (!parsed.success) {
+      return NextResponse.json(
+        { error: parsed.error.issues.map((issue) => issue.message).join(', ') },
+        { status: 400 },
+      )
+    }
+
+    const baseline = setEvalBaseline(parsed.data)
+    const gate = evaluateEvalGate({
+      agentId: parsed.data.agentId,
+      scenarioId: parsed.data.scenarioId,
+      suite: parsed.data.suite,
+      minPercent: parsed.data.minPercent,
+      maxRegressionPoints: parsed.data.maxRegressionPoints,
+    })
+    return NextResponse.json({ baseline, gate })
+  } catch (err: unknown) {
+    return NextResponse.json(
+      { error: errorMessage(err) },
+      { status: 500 },
+    )
+  }
+}

package/src/app/api/eval/gate/route.ts

@@ -0,0 +1,36 @@
+import { NextResponse } from 'next/server'
+import { evaluateEvalGate } from '@/lib/server/eval/baseline'
+import { errorMessage } from '@/lib/shared-utils'
+
+function parseNumberParam(value: string | null): number | null {
+  if (value == null || value.trim() === '') return null
+  const parsed = Number(value)
+  return Number.isFinite(parsed) ? parsed : null
+}
+
+export async function GET(req: Request) {
+  try {
+    const { searchParams } = new URL(req.url)
+    const agentId = searchParams.get('agentId') || ''
+    if (!agentId) {
+      return NextResponse.json(
+        { error: 'agentId is required' },
+        { status: 400 },
+      )
+    }
+
+    const result = evaluateEvalGate({
+      agentId,
+      scenarioId: searchParams.get('scenarioId'),
+      suite: searchParams.get('suite'),
+      minPercent: parseNumberParam(searchParams.get('minPercent')),
+      maxRegressionPoints: parseNumberParam(searchParams.get('maxRegressionPoints')),
+    })
+    return NextResponse.json(result)
+  } catch (err: unknown) {
+    return NextResponse.json(
+      { error: errorMessage(err) },
+      { status: 500 },
+    )
+  }
+}

package/src/cli/index.js

@@ -232,9 +232,12 @@ const COMMAND_GROUPS = [
       cmd('suites', 'GET', '/eval/suites', 'List available eval suites (core, swe-bench-lite, gaia-l1, ...)'),
       cmd('status', 'GET', '/eval/run', 'Get eval run status'),
       cmd('environment', 'GET', '/eval/environments', 'Preview validation environment readiness for an eval'),
+      cmd('baselines', 'GET', '/eval/baselines', 'List eval regression baselines'),
+      cmd('gate', 'GET', '/eval/gate', 'Check the latest eval score against thresholds and baseline'),
       cmd('run', 'POST', '/eval/run', 'Run an eval scenario against an agent', { expectsJsonBody: true }),
       cmd('suite', 'POST', '/eval/suite', 'Run a full eval suite against an agent (pass { suite: "swe-bench-lite" } in body)', { expectsJsonBody: true }),
       cmd('environment-prepare', 'POST', '/eval/environments', 'Prepare validation environment readiness for an eval', { expectsJsonBody: true }),
+      cmd('baseline-set', 'POST', '/eval/baselines', 'Set an eval regression baseline from latest completed runs', { expectsJsonBody: true }),
     ],
   },
   {

package/src/components/quality/quality-workspace.tsx

@@ -17,7 +17,7 @@ import {
 } from '@/lib/quality/quality-summary'
 import { cn } from '@/lib/utils'
 import { useAppStore } from '@/stores/use-app-store'
-import type { EvalEnvironmentPlan, EvalRun, EvalSuiteResult } from '@/lib/server/eval/types'
+import type { EvalEnvironmentPlan, EvalGateResult, EvalRun, EvalSuiteResult } from '@/lib/server/eval/types'
 import type { Agent, ApprovalRequest, SessionRunRecord } from '@/types'
 
 type QualityTab = 'overview' | 'evals' | 'approvals' | 'runs'
@@ -117,6 +117,18 @@ function checkClass(level: 'info' | 'warn' | 'error'): string {
   return 'border-white/[0.06] bg-white/[0.025] text-text-3'
 }
 
+function gateStatusClass(status: EvalGateResult['status']): string {
+  if (status === 'pass') return 'border-emerald-500/25 bg-emerald-500/10 text-emerald-200'
+  if (status === 'warn') return 'border-amber-500/25 bg-amber-500/10 text-amber-200'
+  return 'border-rose-500/25 bg-rose-500/10 text-rose-200'
+}
+
+function gateCheckClass(status: EvalGateResult['status']): string {
+  if (status === 'fail') return 'border-rose-500/20 bg-rose-500/[0.05] text-rose-200'
+  if (status === 'warn') return 'border-amber-500/20 bg-amber-500/[0.05] text-amber-200'
+  return 'border-emerald-500/20 bg-emerald-500/[0.05] text-emerald-200'
+}
+
 function EvalEnvironmentPanel({ plan, loading, onRefresh }: {
   plan: EvalEnvironmentPlan | null
   loading: boolean
@@ -195,6 +207,115 @@ function EvalEnvironmentPanel({ plan, loading, onRefresh }: {
   )
 }
 
+function EvalGatePanel({
+  gate,
+  loading,
+  busy,
+  scope,
+  onScopeChange,
+  onRefresh,
+  onSetBaseline,
+}: {
+  gate: EvalGateResult | null
+  loading: boolean
+  busy: boolean
+  scope: 'scenario' | 'suite'
+  onScopeChange: (scope: 'scenario' | 'suite') => void
+  onRefresh: () => void
+  onSetBaseline: () => void
+}) {
+  return (
+    <div className="rounded-[12px] border border-white/[0.06] bg-white/[0.025] px-3 py-3">
+      <div className="flex items-start justify-between gap-3">
+        <div>
+          <div className="text-[13px] font-800 text-text">Regression gate</div>
+          <p className="mt-1 text-[11px] leading-relaxed text-text-3/65">
+            Compare latest eval evidence against thresholds and an approved baseline.
+          </p>
+        </div>
+        <button
+          type="button"
+          onClick={onRefresh}
+          disabled={loading}
+          className="shrink-0 rounded-[8px] border border-white/[0.08] px-2 py-1 text-[10px] font-800 text-text-2 transition-colors hover:bg-white/[0.06] disabled:opacity-40"
+        >
+          {loading ? 'Checking' : 'Refresh'}
+        </button>
+      </div>
+
+      <div className="mt-3 flex rounded-[10px] border border-white/[0.06] bg-white/[0.025] p-1">
+        {(['scenario', 'suite'] as const).map((item) => (
+          <button
+            key={item}
+            type="button"
+            onClick={() => onScopeChange(item)}
+            className={cn(
+              'flex-1 rounded-[8px] px-2 py-1.5 text-[10px] font-800 uppercase tracking-[0.08em] transition-colors',
+              scope === item ? 'bg-white/[0.1] text-text' : 'text-text-3 hover:bg-white/[0.05]',
+            )}
+          >
+            {item}
+          </button>
+        ))}
+      </div>
+
+      {!gate ? (
+        <div className="mt-3 text-[11px] text-text-3/60">{loading ? 'Checking gate...' : 'Run evals to build gate evidence.'}</div>
+      ) : (
+        <div className="mt-3 flex flex-col gap-3">
+          <div className="flex flex-wrap items-center gap-2">
+            <span className={cn('rounded-full border px-2 py-1 text-[10px] font-800 uppercase tracking-[0.08em]', gateStatusClass(gate.status))}>
+              {gate.status}
+            </span>
+            <span className="rounded-full bg-white/[0.05] px-2 py-1 text-[10px] font-700 text-text-3">
+              {gate.scope.label}
+            </span>
+            <span className="rounded-full bg-white/[0.05] px-2 py-1 text-[10px] font-700 text-text-3">
+              {gate.latestRuns.length}/{gate.scope.scenarioIds.length} latest runs
+            </span>
+          </div>
+
+          <div className="grid grid-cols-3 gap-2">
+            <div className="rounded-[10px] border border-white/[0.06] bg-white/[0.02] px-2 py-2">
+              <div className="text-[9px] font-800 uppercase tracking-[0.08em] text-text-3/50">Current</div>
+              <div className="mt-1 text-[14px] font-800 text-text">{formatPercent(gate.currentPercent)}</div>
+            </div>
+            <div className="rounded-[10px] border border-white/[0.06] bg-white/[0.02] px-2 py-2">
+              <div className="text-[9px] font-800 uppercase tracking-[0.08em] text-text-3/50">Baseline</div>
+              <div className="mt-1 text-[14px] font-800 text-text">{gate.baseline ? `${gate.baseline.baselinePercent}%` : 'none'}</div>
+            </div>
+            <div className="rounded-[10px] border border-white/[0.06] bg-white/[0.02] px-2 py-2">
+              <div className="text-[9px] font-800 uppercase tracking-[0.08em] text-text-3/50">Regression</div>
+              <div className="mt-1 text-[14px] font-800 text-text">{gate.regressionPoints == null ? 'n/a' : `${gate.regressionPoints}pt`}</div>
+            </div>
+          </div>
+
+          <div className="flex flex-col gap-1.5">
+            {gate.checks.slice(0, 4).map((check) => (
+              <div key={`${check.code}:${check.message}`} className={cn('rounded-[9px] border px-2.5 py-2 text-[11px] leading-relaxed', gateCheckClass(check.status))}>
+                <span className="font-800 uppercase tracking-[0.08em]">{check.status}</span>
+                <span className="ml-2">{check.message}</span>
+              </div>
+            ))}
+            {gate.checks.length > 4 && (
+              <div className="text-[10px] text-text-3/55">+{gate.checks.length - 4} more check{gate.checks.length - 4 === 1 ? '' : 's'}</div>
+            )}
+          </div>
+
+          <button
+            type="button"
+            onClick={onSetBaseline}
+            disabled={busy || gate.latestRuns.length === 0 || gate.checks.some((check) => check.code === 'missing_scope_runs')}
+            className="rounded-[9px] border border-white/[0.08] bg-white/[0.04] px-3 py-2 text-[11px] font-800 text-text-2 transition-colors hover:bg-white/[0.08] disabled:cursor-not-allowed disabled:opacity-40"
+          >
+            {busy ? 'Saving baseline' : gate.baseline ? 'Update baseline' : 'Set baseline'}
+          </button>
+        </div>
+      )}
+    </div>
+  )
+}
+
 export function QualityWorkspace() {
   const router = useRouter()
   const searchParams = useSearchParams()
@@ -219,6 +340,10 @@ export function QualityWorkspace() {
   const [evalBusy, setEvalBusy] = useState<string | null>(null)
   const [evalEnvironmentPlan, setEvalEnvironmentPlan] = useState<EvalEnvironmentPlan | null>(null)
   const [evalEnvironmentLoading, setEvalEnvironmentLoading] = useState(false)
+  const [evalGate, setEvalGate] = useState<EvalGateResult | null>(null)
+  const [evalGateScope, setEvalGateScope] = useState<'scenario' | 'suite'>('scenario')
+  const [evalGateLoading, setEvalGateLoading] = useState(false)
+  const [evalBaselineBusy, setEvalBaselineBusy] = useState(false)
   const [approvalBusy, setApprovalBusy] = useState<string | null>(null)
 
   useEffect(() => {
@@ -283,6 +408,30 @@ export function QualityWorkspace() {
     }
   }, [selectedAgentId, selectedScenarioId, selectedSuite])
 
+  const loadEvalGate = useCallback(async () => {
+    if (!selectedAgentId) {
+      setEvalGate(null)
+      return
+    }
+    if (evalGateScope === 'scenario' && !selectedScenarioId) {
+      setEvalGate(null)
+      return
+    }
+    const params = new URLSearchParams({ agentId: selectedAgentId })
+    if (evalGateScope === 'scenario') params.set('scenarioId', selectedScenarioId)
+    else params.set('suite', selectedSuite)
+    setEvalGateLoading(true)
+    try {
+      const gate = await api<EvalGateResult>('GET', `/eval/gate?${params.toString()}`)
+      setEvalGate(gate)
+    } catch (err) {
+      setEvalGate(null)
+      toast.error(err instanceof Error ? err.message : 'Unable to check eval gate')
+    } finally {
+      setEvalGateLoading(false)
+    }
+  }, [evalGateScope, selectedAgentId, selectedScenarioId, selectedSuite])
+
   useEffect(() => {
     void loadQualityData()
   }, [loadQualityData])
@@ -301,6 +450,10 @@ export function QualityWorkspace() {
     void loadEvalEnvironmentPlan()
   }, [loadEvalEnvironmentPlan])
 
+  useEffect(() => {
+    void loadEvalGate()
+  }, [loadEvalGate])
+
   useEffect(() => {
     if (!suites.some((suite) => suite.name === selectedSuite) && suites[0]) {
       setSelectedSuite(suites[0].name)
@@ -341,12 +494,13 @@ export function QualityWorkspace() {
       toast.success('Eval scenario completed')
       await loadQualityData({ silent: true })
       await loadEvalEnvironmentPlan()
+      await loadEvalGate()
     } catch (err) {
       toast.error(err instanceof Error ? err.message : 'Eval scenario failed')
     } finally {
       setEvalBusy(null)
     }
-  }, [evalEnvironmentPlan, loadEvalEnvironmentPlan, loadQualityData, selectedAgentId, selectedScenarioId])
+  }, [evalEnvironmentPlan, loadEvalEnvironmentPlan, loadEvalGate, loadQualityData, selectedAgentId, selectedScenarioId])
 
   const runSuite = useCallback(async (suiteName: string) => {
     if (!selectedAgentId) {
@@ -369,12 +523,37 @@ export function QualityWorkspace() {
       toast.success(`Suite completed at ${Math.round(result.percentage)}%`)
       await loadQualityData({ silent: true })
       await loadEvalEnvironmentPlan()
+      await loadEvalGate()
     } catch (err) {
       toast.error(err instanceof Error ? err.message : 'Eval suite failed')
     } finally {
       setEvalBusy(null)
     }
-  }, [evalEnvironmentPlan, loadEvalEnvironmentPlan, loadQualityData, selectedAgentId])
+  }, [evalEnvironmentPlan, loadEvalEnvironmentPlan, loadEvalGate, loadQualityData, selectedAgentId])
+
+  const setEvalBaseline = useCallback(async () => {
+    if (!selectedAgentId) {
+      toast.error('Choose an agent first')
+      return
+    }
+    if (evalGateScope === 'scenario' && !selectedScenarioId) {
+      toast.error('Choose a scenario first')
+      return
+    }
+    setEvalBaselineBusy(true)
+    try {
+      const body = evalGateScope === 'scenario'
+        ? { agentId: selectedAgentId, scenarioId: selectedScenarioId, minPercent: evalGate?.minPercent ?? 80, maxRegressionPoints: evalGate?.maxRegressionPoints ?? 5 }
+        : { agentId: selectedAgentId, suite: selectedSuite, minPercent: evalGate?.minPercent ?? 80, maxRegressionPoints: evalGate?.maxRegressionPoints ?? 5 }
+      const result = await api<{ gate: EvalGateResult }>('POST', '/eval/baselines', body)
+      setEvalGate(result.gate)
+      toast.success('Eval baseline saved')
+    } catch (err) {
+      toast.error(err instanceof Error ? err.message : 'Unable to save eval baseline')
+    } finally {
+      setEvalBaselineBusy(false)
+    }
+  }, [evalGate, evalGateScope, selectedAgentId, selectedScenarioId, selectedSuite])
 
   const actOnApproval = useCallback(async (approval: ApprovalRequest, approved: boolean) => {
     setApprovalBusy(approval.id)
@@ -600,6 +779,15 @@ export function QualityWorkspace() {
                     loading={evalEnvironmentLoading}
                     onRefresh={() => void loadEvalEnvironmentPlan({ refreshGateway: true })}
                   />
+                  <EvalGatePanel
+                    gate={evalGate}
+                    loading={evalGateLoading}
+                    busy={evalBaselineBusy}
+                    scope={evalGateScope}
+                    onScopeChange={setEvalGateScope}
+                    onRefresh={() => void loadEvalGate()}
+                    onSetBaseline={() => void setEvalBaseline()}
+                  />
                   <button
                     type="button"
                     onClick={() => openMissionTemplate('release-candidate-qa')}

package/src/lib/server/eval/baseline.test.ts

@@ -0,0 +1,111 @@
+import assert from 'node:assert/strict'
+import test from 'node:test'
+
+import {
+  evaluateEvalGate,
+  setEvalBaseline,
+} from './baseline'
+import type { EvalBaseline, EvalRun } from './types'
+
+function makeRun(overrides: Partial<EvalRun> = {}): EvalRun {
+  return {
+    id: 'run-1',
+    scenarioId: 'coding-prime',
+    agentId: 'agent-1',
+    status: 'completed',
+    startedAt: 1,
+    endedAt: 2,
+    score: 8,
+    maxScore: 10,
+    details: [],
+    ...overrides,
+  }
+}
+
+function depsFor(runs: EvalRun[], baseline: EvalBaseline | null = null, saved: EvalBaseline[] = []) {
+  return {
+    now: () => 123,
+    listRunsByAgent: (agentId: string) => runs.filter((run) => run.agentId === agentId),
+    getBaselineForScope: () => baseline,
+    saveBaseline: (next: EvalBaseline) => { saved.push(next) },
+  }
+}
+
+test('setEvalBaseline snapshots the latest scenario score and gate defaults', () => {
+  const saved: EvalBaseline[] = []
+  const baseline = setEvalBaseline(
+    {
+      agentId: 'agent-1',
+      scenarioId: 'coding-prime',
+      minPercent: 75,
+      maxRegressionPoints: 3,
+      label: 'Release candidate',
+    },
+    depsFor([
+      makeRun({ id: 'older', score: 4, startedAt: 1, endedAt: 2 }),
+      makeRun({ id: 'latest', score: 8, startedAt: 5, endedAt: 6 }),
+    ], null, saved),
+  )
+
+  assert.equal(saved.length, 1)
+  assert.equal(baseline.scope.type, 'scenario')
+  assert.equal(baseline.scope.id, 'coding-prime')
+  assert.equal(baseline.baselinePercent, 80)
+  assert.equal(baseline.minPercent, 75)
+  assert.equal(baseline.maxRegressionPoints, 3)
+  assert.deepEqual(baseline.runIds, ['latest'])
+})
+
+test('evaluateEvalGate warns until a baseline is approved', () => {
+  const gate = evaluateEvalGate(
+    { agentId: 'agent-1', scenarioId: 'coding-prime', minPercent: 70 },
+    depsFor([makeRun({ score: 8, maxScore: 10 })]),
+  )
+
+  assert.equal(gate.currentPercent, 80)
+  assert.equal(gate.status, 'warn')
+  assert.ok(gate.checks.some((check) => check.code === 'baseline_missing' && check.status === 'warn'))
+})
+
+test('evaluateEvalGate fails when regression exceeds the baseline allowance', () => {
+  const baseline = setEvalBaseline(
+    { agentId: 'agent-1', scenarioId: 'coding-prime', minPercent: 70, maxRegressionPoints: 2 },
+    depsFor([makeRun({ id: 'baseline', score: 9, maxScore: 10 })]),
+  )
+
+  const gate = evaluateEvalGate(
+    { agentId: 'agent-1', scenarioId: 'coding-prime' },
+    depsFor([makeRun({ id: 'current', score: 6, maxScore: 10, startedAt: 10, endedAt: 11 })], baseline),
+  )
+
+  assert.equal(gate.currentPercent, 60)
+  assert.equal(gate.regressionPoints, 30)
+  assert.equal(gate.status, 'fail')
+  assert.ok(gate.checks.some((check) => check.code === 'regression_limit_exceeded'))
+})
+
+test('evaluateEvalGate passes when score and regression checks pass', () => {
+  const baseline = setEvalBaseline(
+    { agentId: 'agent-1', scenarioId: 'coding-prime', minPercent: 70, maxRegressionPoints: 5 },
+    depsFor([makeRun({ id: 'baseline', score: 8, maxScore: 10 })]),
+  )
+
+  const gate = evaluateEvalGate(
+    { agentId: 'agent-1', scenarioId: 'coding-prime' },
+    depsFor([makeRun({ id: 'current', score: 8, maxScore: 10, startedAt: 10, endedAt: 11 })], baseline),
+  )
+
+  assert.equal(gate.status, 'pass')
+  assert.equal(gate.regressionPoints, 0)
+  assert.ok(gate.checks.some((check) => check.code === 'score_threshold_met'))
+})
+
+test('suite gates require latest runs for every scenario in scope before baselining', () => {
+  assert.throws(
+    () => setEvalBaseline(
+      { agentId: 'agent-1', suite: 'core' },
+      depsFor([makeRun({ scenarioId: 'coding-prime' })]),
+    ),
+    /Baseline requires latest runs for every scenario in scope/,
+  )
+})

package/src/lib/server/eval/baseline.ts

@@ -0,0 +1,274 @@
+import {
+  getEvalBaselineForScope,
+  listEvalBaselines,
+  listEvalRunsByAgent,
+  saveEvalBaseline,
+} from './store'
+import { getScenario, getSuiteScenarios } from './scenarios'
+import type {
+  EvalBaseline,
+  EvalGateCheck,
+  EvalGateResult,
+  EvalGateScope,
+  EvalGateScopeType,
+  EvalRun,
+} from './types'
+
+const DEFAULT_MIN_PERCENT = 80
+const DEFAULT_MAX_REGRESSION_POINTS = 5
+const MAX_LOOKBACK_RUNS = 1_000
+
+export interface EvalGateInput {
+  agentId: string
+  scenarioId?: string | null
+  suite?: string | null
+  minPercent?: number | null
+  maxRegressionPoints?: number | null
+}
+
+export interface SetEvalBaselineInput extends EvalGateInput {
+  label?: string | null
+  notes?: string | null
+}
+
+interface EvalGateDeps {
+  now?: () => number
+  listRunsByAgent?: (agentId: string, limit: number) => EvalRun[]
+  getBaselineForScope?: (agentId: string, scopeType: EvalGateScopeType, scopeId: string) => EvalBaseline | null
+  saveBaseline?: (baseline: EvalBaseline) => void
+  listBaselines?: (filters?: { agentId?: string; limit?: number }) => EvalBaseline[]
+}
+
+interface EvalAggregate {
+  runs: EvalRun[]
+  missingScenarioIds: string[]
+  score: number
+  maxScore: number
+  percent: number | null
+}
+
+function normalizePercent(value: number | null | undefined, fallback: number): number {
+  if (!Number.isFinite(value) || value == null) return fallback
+  return Math.max(0, Math.min(100, Math.round(value)))
+}
+
+function normalizeRegressionPoints(value: number | null | undefined, fallback: number): number {
+  if (!Number.isFinite(value) || value == null) return fallback
+  return Math.max(0, Math.round(value))
+}
+
+function scorePercent(score: number, maxScore: number): number | null {
+  if (!Number.isFinite(score) || !Number.isFinite(maxScore) || maxScore <= 0) return null
+  return Math.round((score / maxScore) * 100)
+}
+
+function maxScoreForScenario(scenarioId: string): number {
+  const scenario = getScenario(scenarioId)
+  return scenario?.scoringCriteria.reduce((sum, criterion) => sum + criterion.weight, 0) ?? 0
+}
+
+export function resolveEvalGateScope(input: Pick<EvalGateInput, 'scenarioId' | 'suite'>): EvalGateScope {
+  const scenarioId = input.scenarioId?.trim()
+  if (scenarioId) {
+    const scenario = getScenario(scenarioId)
+    if (!scenario) throw new Error(`Unknown eval scenario: ${scenarioId}`)
+    return {
+      type: 'scenario',
+      id: scenario.id,
+      label: scenario.name,
+      scenarioIds: [scenario.id],
+    }
+  }
+
+  const suite = input.suite?.trim() || 'core'
+  const scenarios = getSuiteScenarios(suite)
+  if (scenarios.length === 0) throw new Error(`Unknown or empty eval suite: ${suite}`)
+  return {
+    type: 'suite',
+    id: suite,
+    label: suite,
+    scenarioIds: scenarios.map((scenario) => scenario.id),
+  }
+}
+
+export function evalBaselineId(agentId: string, scope: EvalGateScope): string {
+  return `eval-baseline:${agentId}:${scope.type}:${scope.id}`
+}
+
+function latestRunsForScope(runs: EvalRun[], scope: EvalGateScope): EvalRun[] {
+  const scenarioSet = new Set(scope.scenarioIds)
+  const latest = new Map<string, EvalRun>()
+
+  for (const run of runs) {
+    if (!scenarioSet.has(run.scenarioId)) continue
+    if (run.status === 'pending' || run.status === 'running') continue
+    const previous = latest.get(run.scenarioId)
+    if (!previous || (run.endedAt ?? run.startedAt) > (previous.endedAt ?? previous.startedAt)) {
+      latest.set(run.scenarioId, run)
+    }
+  }
+
+  return scope.scenarioIds
+    .map((scenarioId) => latest.get(scenarioId))
+    .filter(Boolean) as EvalRun[]
+}
+
+function aggregateRuns(scope: EvalGateScope, runs: EvalRun[]): EvalAggregate {
+  const byScenario = new Map(runs.map((run) => [run.scenarioId, run]))
+  const missingScenarioIds = scope.scenarioIds.filter((scenarioId) => !byScenario.has(scenarioId))
+  const score = scope.scenarioIds.reduce((sum, scenarioId) => sum + (byScenario.get(scenarioId)?.score ?? 0), 0)
+  const maxScore = scope.scenarioIds.reduce((sum, scenarioId) => {
+    const runMaxScore = byScenario.get(scenarioId)?.maxScore
+    return sum + (Number.isFinite(runMaxScore) && runMaxScore != null ? runMaxScore : maxScoreForScenario(scenarioId))
+  }, 0)
+  return {
+    runs,
+    missingScenarioIds,
+    score,
+    maxScore,
+    percent: scorePercent(score, maxScore),
+  }
+}
+
+function statusFromChecks(checks: EvalGateCheck[]): EvalGateResult['status'] {
+  if (checks.some((check) => check.status === 'fail')) return 'fail'
+  if (checks.some((check) => check.status === 'warn')) return 'warn'
+  return 'pass'
+}
+
+export function listEvalBaselinesForAgent(agentId?: string | null, deps: EvalGateDeps = {}): EvalBaseline[] {
+  const list = deps.listBaselines || listEvalBaselines
+  return list({ agentId: agentId || undefined, limit: 200 })
+}
+
+export function setEvalBaseline(input: SetEvalBaselineInput, deps: EvalGateDeps = {}): EvalBaseline {
+  if (!input.agentId.trim()) throw new Error('agentId is required')
+
+  const now = deps.now?.() ?? Date.now()
+  const scope = resolveEvalGateScope(input)
+  const runs = latestRunsForScope(
+    (deps.listRunsByAgent || listEvalRunsByAgent)(input.agentId, MAX_LOOKBACK_RUNS),
+    scope,
+  )
+  const aggregate = aggregateRuns(scope, runs)
+  if (aggregate.runs.length === 0) {
+    throw new Error('Run the selected eval before setting a baseline.')
+  }
+  if (aggregate.missingScenarioIds.length > 0) {
+    throw new Error(`Baseline requires latest runs for every scenario in scope. Missing: ${aggregate.missingScenarioIds.join(', ')}`)
+  }
+
+  const existing = (deps.getBaselineForScope || getEvalBaselineForScope)(input.agentId, scope.type, scope.id)
+  const baseline: EvalBaseline = {
+    id: existing?.id || evalBaselineId(input.agentId, scope),
+    agentId: input.agentId,
+    scope,
+    baselineScore: aggregate.score,
+    baselineMaxScore: aggregate.maxScore,
+    baselinePercent: aggregate.percent ?? 0,
+    minPercent: normalizePercent(input.minPercent, aggregate.percent ?? DEFAULT_MIN_PERCENT),
+    maxRegressionPoints: normalizeRegressionPoints(input.maxRegressionPoints, existing?.maxRegressionPoints ?? DEFAULT_MAX_REGRESSION_POINTS),
+    runIds: aggregate.runs.map((run) => run.id),
+    label: input.label?.trim() || existing?.label || null,
+    notes: input.notes?.trim() || existing?.notes || null,
+    createdAt: existing?.createdAt || now,
+    updatedAt: now,
+  }
+
+  ;(deps.saveBaseline || saveEvalBaseline)(baseline)
+  return baseline
+}
+
+export function evaluateEvalGate(input: EvalGateInput, deps: EvalGateDeps = {}): EvalGateResult {
+  if (!input.agentId.trim()) throw new Error('agentId is required')
+
+  const generatedAt = deps.now?.() ?? Date.now()
+  const scope = resolveEvalGateScope(input)
+  const baseline = (deps.getBaselineForScope || getEvalBaselineForScope)(input.agentId, scope.type, scope.id)
+  const runs = latestRunsForScope(
+    (deps.listRunsByAgent || listEvalRunsByAgent)(input.agentId, MAX_LOOKBACK_RUNS),
+    scope,
+  )
+  const aggregate = aggregateRuns(scope, runs)
+  const minPercent = normalizePercent(input.minPercent, baseline?.minPercent ?? DEFAULT_MIN_PERCENT)
+  const maxRegressionPoints = normalizeRegressionPoints(input.maxRegressionPoints, baseline?.maxRegressionPoints ?? DEFAULT_MAX_REGRESSION_POINTS)
+  const regressionPoints = baseline && aggregate.percent != null
+    ? Math.max(0, baseline.baselinePercent - aggregate.percent)
+    : null
+
+  const checks: EvalGateCheck[] = []
+  if (aggregate.runs.length === 0) {
+    checks.push({
+      code: 'no_eval_runs',
+      status: 'fail',
+      message: 'No completed eval runs are available for this gate.',
+    })
+  }
+  if (aggregate.missingScenarioIds.length > 0) {
+    checks.push({
+      code: 'missing_scope_runs',
+      status: 'fail',
+      message: `${aggregate.missingScenarioIds.length} scenario${aggregate.missingScenarioIds.length === 1 ? '' : 's'} have no latest run in this gate.`,
+      detail: aggregate.missingScenarioIds.join(', '),
+    })
+  }
+  if (aggregate.runs.some((run) => run.status === 'failed')) {
+    checks.push({
+      code: 'failed_eval_run',
+      status: 'fail',
+      message: 'At least one latest eval run failed.',
+    })
+  }
+  if (aggregate.percent == null || aggregate.percent < minPercent) {
+    checks.push({
+      code: 'score_below_threshold',
+      status: 'fail',
+      message: `Current score is below the ${minPercent}% gate.`,
+      detail: aggregate.percent == null ? 'n/a' : `${aggregate.percent}%`,
+    })
+  } else {
+    checks.push({
+      code: 'score_threshold_met',
+      status: 'pass',
+      message: `Current score meets the ${minPercent}% gate.`,
+      detail: `${aggregate.percent}%`,
+    })
+  }
+  if (!baseline) {
+    checks.push({
+      code: 'baseline_missing',
+      status: 'warn',
+      message: 'No approved baseline is set for this gate.',
+    })
+  } else if (regressionPoints != null && regressionPoints > maxRegressionPoints) {
+    checks.push({
+      code: 'regression_limit_exceeded',
+      status: 'fail',
+      message: `Regression exceeds the ${maxRegressionPoints} point allowance.`,
+      detail: `${regressionPoints} points below baseline`,
+    })
+  } else if (regressionPoints != null) {
+    checks.push({
+      code: 'regression_within_limit',
+      status: 'pass',
+      message: `Regression is within the ${maxRegressionPoints} point allowance.`,
+      detail: `${regressionPoints} point${regressionPoints === 1 ? '' : 's'} below baseline`,
+    })
+  }
+
+  return {
+    agentId: input.agentId,
+    scope,
+    status: statusFromChecks(checks),
+    generatedAt,
+    baseline,
+    latestRuns: aggregate.runs,
+    currentScore: aggregate.score,
+    currentMaxScore: aggregate.maxScore,
+    currentPercent: aggregate.percent,
+    regressionPoints,
+    minPercent,
+    maxRegressionPoints,
+    checks,
+  }
+}

package/src/lib/server/eval/store.ts

@@ -1,6 +1,6 @@
 import Database from 'better-sqlite3'
 import path from 'path'
-import type { EvalRun } from './types'
+import type { EvalBaseline, EvalRun } from './types'
 import { DATA_DIR } from '../data-dir'
 
 const DB_PATH = path.join(DATA_DIR, 'eval-runs.db')
@@ -15,6 +15,15 @@ function getDb(): Database.Database {
       id TEXT PRIMARY KEY,
       data TEXT NOT NULL
     )`)
+    db.exec(`CREATE TABLE IF NOT EXISTS eval_baselines (
+      id TEXT PRIMARY KEY,
+      agent_id TEXT NOT NULL,
+      scope_type TEXT NOT NULL,
+      scope_id TEXT NOT NULL,
+      data TEXT NOT NULL,
+      updated_at INTEGER NOT NULL
+    )`)
+    db.exec('CREATE INDEX IF NOT EXISTS idx_eval_baselines_agent ON eval_baselines(agent_id, scope_type, scope_id)')
   }
   return db
 }
@@ -36,3 +45,40 @@ export function listEvalRuns(limit = 50): EvalRun[] {
 export function listEvalRunsByAgent(agentId: string, limit = 50): EvalRun[] {
   return listEvalRuns(limit * 2).filter(r => r.agentId === agentId).slice(0, limit)
 }
+
+export function saveEvalBaseline(baseline: EvalBaseline): void {
+  getDb().prepare(`
+    INSERT OR REPLACE INTO eval_baselines (id, agent_id, scope_type, scope_id, data, updated_at)
+    VALUES (?, ?, ?, ?, ?, ?)
+  `).run(
+    baseline.id,
+    baseline.agentId,
+    baseline.scope.type,
+    baseline.scope.id,
+    JSON.stringify(baseline),
+    baseline.updatedAt,
+  )
+}
+
+export function getEvalBaseline(id: string): EvalBaseline | null {
+  const row = getDb().prepare('SELECT data FROM eval_baselines WHERE id = ?').get(id) as { data: string } | undefined
+  return row ? JSON.parse(row.data) as EvalBaseline : null
+}
+
+export function getEvalBaselineForScope(agentId: string, scopeType: EvalBaseline['scope']['type'], scopeId: string): EvalBaseline | null {
+  const row = getDb().prepare(`
+    SELECT data FROM eval_baselines
+    WHERE agent_id = ? AND scope_type = ? AND scope_id = ?
+    ORDER BY updated_at DESC
+    LIMIT 1
+  `).get(agentId, scopeType, scopeId) as { data: string } | undefined
+  return row ? JSON.parse(row.data) as EvalBaseline : null
+}
+
+export function listEvalBaselines(filters: { agentId?: string; limit?: number } = {}): EvalBaseline[] {
+  const limit = Math.max(1, Math.min(filters.limit ?? 100, 500))
+  const rows = filters.agentId
+    ? getDb().prepare('SELECT data FROM eval_baselines WHERE agent_id = ? ORDER BY updated_at DESC LIMIT ?').all(filters.agentId, limit) as { data: string }[]
+    : getDb().prepare('SELECT data FROM eval_baselines ORDER BY updated_at DESC LIMIT ?').all(limit) as { data: string }[]
+  return rows.map((row) => JSON.parse(row.data) as EvalBaseline)
+}

package/src/lib/server/eval/types.ts

@@ -105,3 +105,53 @@ export interface EvalSuiteResult {
   runs: EvalRun[]
   completedAt: number
 }
+
+export type EvalGateScopeType = 'scenario' | 'suite'
+
+export interface EvalGateScope {
+  type: EvalGateScopeType
+  id: string
+  label: string
+  scenarioIds: string[]
+}
+
+export interface EvalBaseline {
+  id: string
+  agentId: string
+  scope: EvalGateScope
+  baselineScore: number
+  baselineMaxScore: number
+  baselinePercent: number
+  minPercent: number
+  maxRegressionPoints: number
+  runIds: string[]
+  label?: string | null
+  notes?: string | null
+  createdAt: number
+  updatedAt: number
+}
+
+export type EvalGateStatus = 'pass' | 'warn' | 'fail'
+
+export interface EvalGateCheck {
+  code: string
+  status: EvalGateStatus
+  message: string
+  detail?: string
+}
+
+export interface EvalGateResult {
+  agentId: string
+  scope: EvalGateScope
+  status: EvalGateStatus
+  generatedAt: number
+  baseline: EvalBaseline | null
+  latestRuns: EvalRun[]
+  currentScore: number
+  currentMaxScore: number
+  currentPercent: number | null
+  regressionPoints: number | null
+  minPercent: number
+  maxRegressionPoints: number
+  checks: EvalGateCheck[]
+}

package/src/lib/server/session-tools/extension-creator.ts

@@ -138,7 +138,7 @@ module.exports = {
     }
   ],
 
-  // --- Managed Resources (Paperclip-compatible) ---
+  // --- Managed Resources ---
   managedResources: {
     agents: [
       {
@@ -211,7 +211,7 @@ Key rules:
 - Dependency installs are run by the extension manager inside a per-extension workspace using the selected package manager with scripts disabled.
 - Extension settings are declared through ui.settingsFields and stored per extension ID
 - Managed resources let an extension declare provisionable agents, schedules/routines, trusted local folders, gateway platforms, and setup checks. Operators reconcile them through Extensions > Managed Resources or /api/extensions/managed-resources.
-- Paperclip-compatible top-level agents, routines, and localFolders are also accepted; SwarmClaw reconciles routines as schedules when they include schedule timing.
+- Top-level agents, routines, and localFolders are also accepted; SwarmClaw reconciles routines as schedules when they include schedule timing.
 - Keep extensions focused: one clear purpose per extension
 `
     }

package/src/lib/server/tasks/task-checkout.ts

@@ -6,7 +6,7 @@ import { loadTasks, saveTasks } from '@/lib/server/tasks/task-repository'
  * Atomically transition a task from queued → running with a checkout run ID.
  *
  * Uses a SQLite IMMEDIATE transaction to prevent two runners from starting the
- * same task concurrently (Paperclip-inspired atomic checkout pattern).
+ * same task concurrently.
  *
  * Returns the checked-out task on success, or null if the task was already
  * taken, missing, or no longer in queued status.

package/src/types/extension.ts

@@ -362,10 +362,10 @@ export interface ExtensionSetupCheckDeclaration {
 export interface ExtensionManagedResources {
   agents?: ExtensionManagedAgentDeclaration[]
   schedules?: ExtensionManagedScheduleDeclaration[]
-  /** Paperclip-compatible alias. SwarmClaw reconciles routines as managed schedules. */
+  /** Routine alias. SwarmClaw reconciles routines as managed schedules. */
   routines?: ExtensionManagedScheduleDeclaration[]
   localFolders?: ExtensionManagedLocalFolderDeclaration[]
-  /** Hermes-style gateway/platform declaration metadata for setup and diagnostics surfaces. */
+  /** Gateway/platform declaration metadata for setup and diagnostics surfaces. */
   gatewayPlatforms?: ExtensionGatewayPlatformDeclaration[]
   setupChecks?: ExtensionSetupCheckDeclaration[]
 }
@@ -420,7 +420,7 @@ export interface Extension {
   providers?: ExtensionProviderDefinition[]
   connectors?: ExtensionConnectorDefinition[]
   managedResources?: ExtensionManagedResources
-  /** Paperclip-compatible top-level aliases. Prefer managedResources for new SwarmClaw extensions. */
+  /** Top-level managed-resource aliases. Prefer managedResources for new SwarmClaw extensions. */
   agents?: ExtensionManagedAgentDeclaration[]
   schedules?: ExtensionManagedScheduleDeclaration[]
   routines?: ExtensionManagedScheduleDeclaration[]

package/electron-dist/main.js

@@ -1,218 +0,0 @@
-"use strict";
-var __createBinding = (this && this.__createBinding) || (Object.create ? (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    var desc = Object.getOwnPropertyDescriptor(m, k);
-    if (!desc || ("get" in desc ? !m.__esModule : desc.writable || desc.configurable)) {
-      desc = { enumerable: true, get: function() { return m[k]; } };
-    }
-    Object.defineProperty(o, k2, desc);
-}) : (function(o, m, k, k2) {
-    if (k2 === undefined) k2 = k;
-    o[k2] = m[k];
-}));
-var __setModuleDefault = (this && this.__setModuleDefault) || (Object.create ? (function(o, v) {
-    Object.defineProperty(o, "default", { enumerable: true, value: v });
-}) : function(o, v) {
-    o["default"] = v;
-});
-var __importStar = (this && this.__importStar) || (function () {
-    var ownKeys = function(o) {
-        ownKeys = Object.getOwnPropertyNames || function (o) {
-            var ar = [];
-            for (var k in o) if (Object.prototype.hasOwnProperty.call(o, k)) ar[ar.length] = k;
-            return ar;
-        };
-        return ownKeys(o);
-    };
-    return function (mod) {
-        if (mod && mod.__esModule) return mod;
-        var result = {};
-        if (mod != null) for (var k = ownKeys(mod), i = 0; i < k.length; i++) if (k[i] !== "default") __createBinding(result, mod, k[i]);
-        __setModuleDefault(result, mod);
-        return result;
-    };
-})();
-var __importDefault = (this && this.__importDefault) || function (mod) {
-    return (mod && mod.__esModule) ? mod : { "default": mod };
-};
-Object.defineProperty(exports, "__esModule", { value: true });
-const electron_1 = require("electron");
-const node_fs_1 = __importDefault(require("node:fs"));
-const node_path_1 = __importDefault(require("node:path"));
-const paths_1 = require("./paths");
-const server_lifecycle_1 = require("./server-lifecycle");
-const menu_1 = require("./menu");
-const DEV_URL_DEFAULT = 'http://127.0.0.1:3456';
-const LOG_TAIL_BYTES = 1500;
-let mainWindow = null;
-let serverHandle = null;
-let serverLogFile = null;
-let isQuitting = false;
-const gotLock = electron_1.app.requestSingleInstanceLock();
-if (!gotLock) {
-    electron_1.app.quit();
-}
-else {
-    electron_1.app.on('second-instance', () => {
-        if (mainWindow) {
-            if (mainWindow.isMinimized())
-                mainWindow.restore();
-            mainWindow.focus();
-        }
-    });
-    electron_1.app.on('ready', () => void onReady());
-    electron_1.app.on('window-all-closed', () => {
-        if (process.platform !== 'darwin')
-            electron_1.app.quit();
-    });
-    electron_1.app.on('activate', () => {
-        if (mainWindow !== null)
-            return;
-        if (serverHandle) {
-            createMainWindow(serverHandle.url);
-        }
-        else if (!electron_1.app.isPackaged) {
-            createMainWindow(process.env.SWARMCLAW_DEV_URL || DEV_URL_DEFAULT);
-        }
-    });
-    electron_1.app.on('before-quit', () => {
-        isQuitting = true;
-    });
-    electron_1.app.on('will-quit', async (event) => {
-        if (!serverHandle)
-            return;
-        event.preventDefault();
-        try {
-            await serverHandle.stop();
-        }
-        finally {
-            serverHandle = null;
-            electron_1.app.exit(0);
-        }
-    });
-}
-async function onReady() {
-    const paths = (0, paths_1.resolveRuntimePaths)();
-    (0, menu_1.buildAppMenu)(paths, () => mainWindow);
-    const iconPath = resolveIconPath();
-    if (process.platform === 'darwin' && iconPath && electron_1.app.dock) {
-        const img = electron_1.nativeImage.createFromPath(iconPath);
-        if (!img.isEmpty())
-            electron_1.app.dock.setIcon(img);
-    }
-    if (!electron_1.app.isPackaged) {
-        const devUrl = process.env.SWARMCLAW_DEV_URL || DEV_URL_DEFAULT;
-        console.log(`[swarmclaw] dev mode, loading ${devUrl}`);
-        createMainWindow(devUrl);
-        return;
-    }
-    serverLogFile = node_path_1.default.join(electron_1.app.getPath('userData'), 'logs', 'server.log');
-    node_fs_1.default.mkdirSync(node_path_1.default.dirname(serverLogFile), { recursive: true });
-    try {
-        serverHandle = await (0, server_lifecycle_1.startEmbeddedServer)({
-            paths,
-            logFile: serverLogFile,
-            onStdout: (c) => process.stdout.write(`[swarmclaw] ${c}`),
-            onStderr: (c) => process.stderr.write(`[swarmclaw] ${c}`),
-            onExit: (code, signal) => {
-                if (!isQuitting) {
-                    console.error(`[swarmclaw] server exited unexpectedly (code=${code}, signal=${signal ?? 'none'})`);
-                    void showServerCrashDialog(code, signal);
-                }
-            },
-        });
-    }
-    catch (err) {
-        await showStartupFailureDialog(err, paths);
-        electron_1.app.exit(1);
-        return;
-    }
-    createMainWindow(serverHandle.url);
-    void Promise.resolve().then(() => __importStar(require('./updater'))).then((m) => m.initAutoUpdater());
-}
-function resolveIconPath() {
-    const candidate = electron_1.app.isPackaged
-        ? node_path_1.default.join(process.resourcesPath, 'icon.png')
-        : node_path_1.default.join(__dirname, '..', 'resources', 'icon.png');
-    return node_fs_1.default.existsSync(candidate) ? candidate : undefined;
-}
-function createMainWindow(startUrl) {
-    const iconPath = resolveIconPath();
-    mainWindow = new electron_1.BrowserWindow({
-        width: 1440,
-        height: 900,
-        minWidth: 1024,
-        minHeight: 640,
-        backgroundColor: '#0b0b0f',
-        show: true,
-        ...(iconPath ? { icon: iconPath } : {}),
-        webPreferences: {
-            contextIsolation: true,
-            nodeIntegration: false,
-            sandbox: false,
-        },
-    });
-    const wc = mainWindow.webContents;
-    if (!electron_1.app.isPackaged)
-        wc.openDevTools({ mode: 'detach' });
-    wc.on('did-start-loading', () => console.log('[swarmclaw] did-start-loading'));
-    wc.on('did-finish-load', () => console.log('[swarmclaw] did-finish-load'));
-    wc.on('did-fail-load', (_e, code, desc, url) => console.error(`[swarmclaw] did-fail-load code=${code} desc=${desc} url=${url}`));
-    wc.on('render-process-gone', (_e, details) => console.error(`[swarmclaw] render-process-gone reason=${details.reason}`));
-    wc.on('unresponsive', () => console.error('[swarmclaw] webContents unresponsive'));
-    mainWindow.on('closed', () => {
-        mainWindow = null;
-    });
-    mainWindow.webContents.setWindowOpenHandler(({ url }) => {
-        if (url.startsWith(startUrl))
-            return { action: 'allow' };
-        void electron_1.shell.openExternal(url);
-        return { action: 'deny' };
-    });
-    void mainWindow.loadURL(startUrl).catch((err) => {
-        console.error('[swarmclaw] loadURL rejected:', err);
-    });
-}
-async function showServerCrashDialog(code, signal) {
-    const buttons = serverLogFile ? ['Open Logs Folder', 'Quit'] : ['Quit'];
-    const quitButtonId = buttons.length - 1;
-    const detail = buildLogDetail(`code=${code ?? 'null'} signal=${signal ?? 'none'}`);
-    const res = await electron_1.dialog.showMessageBox({
-        type: 'error',
-        buttons,
-        defaultId: quitButtonId,
-        cancelId: quitButtonId,
-        title: 'SwarmClaw stopped',
-        message: 'The SwarmClaw server exited unexpectedly.',
-        detail,
-    });
-    if (serverLogFile && res.response === 0)
-        electron_1.shell.showItemInFolder(serverLogFile);
-    electron_1.app.exit(1);
-}
-async function showStartupFailureDialog(err, paths) {
-    const message = err instanceof Error ? err.message : String(err);
-    const base = `${message}\n\nStandalone entry: ${paths.standaloneEntry}\nData dir: ${paths.dataDir}`;
-    const detail = buildLogDetail(base);
-    const buttons = serverLogFile ? ['Open Logs Folder', 'Quit'] : ['Quit'];
-    const quitButtonId = buttons.length - 1;
-    const res = await electron_1.dialog.showMessageBox({
-        type: 'error',
-        buttons,
-        defaultId: quitButtonId,
-        cancelId: quitButtonId,
-        title: 'SwarmClaw failed to start',
-        message: 'The embedded server did not start.',
-        detail,
-    });
-    if (serverLogFile && res.response === 0)
-        electron_1.shell.showItemInFolder(serverLogFile);
-}
-function buildLogDetail(base) {
-    if (!serverLogFile)
-        return base;
-    const tail = (0, server_lifecycle_1.tailLogFile)(serverLogFile, LOG_TAIL_BYTES).trim();
-    if (!tail)
-        return `${base}\n\nLog file: ${serverLogFile}\n(no output captured yet)`;
-    return `${base}\n\nLog tail (${serverLogFile}):\n${tail}`;
-}