Merge branch 'main' into fix/publish-runtime-autobump-fetch-depth

2026-05-11 11:39:14 +00:00 · 2026-05-11 11:39:14 +00:00 · e0bbba801e
commit e0bbba801e
parent dd992fcc9b 5c10ee0d73
17 changed files with 249 additions and 151 deletions
--- a/.gitea/workflows/continuous-synth-e2e.yml
+++ b/.gitea/workflows/continuous-synth-e2e.yml
@ -56,7 +56,7 @@ on:
    #   2. Avoid colliding with the existing :15 sweep-cf-orphans
    #      and :45 sweep-cf-tunnels — both hit the CF API and we
    #      don't want to fight for rate-limit tokens.
-    #   3. Avoid the :30 heavy slot (canary-staging /30, sweep-aws-
+    #   3. Avoid the :30 heavy slot (staging-smoke /30, sweep-aws-
    #      secrets, sweep-stale-e2e-orgs every :15) — multiple
    #      overlapping cron registrations on the same minute is part
    #      of what GH drops under load.
--- a/.gitea/workflows/e2e-staging-canvas.yml
+++ b/.gitea/workflows/e2e-staging-canvas.yml
@ -124,7 +124,10 @@ jobs:
    env:
      CANVAS_E2E_STAGING: '1'
      MOLECULE_CP_URL: https://staging-api.moleculesai.app
-      MOLECULE_ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+      # 2026-05-11: secret canonicalised from MOLECULE_STAGING_ADMIN_TOKEN
+      # (dead in org secret store) to CP_STAGING_ADMIN_API_TOKEN per
+      # internal#322 — see this PR for the cross-workflow sweep.
+      MOLECULE_ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}

    defaults:
      run:
@ -145,7 +148,7 @@ jobs:
        if: needs.detect-changes.outputs.canvas == 'true'
        run: |
          if [ -z "$MOLECULE_ADMIN_TOKEN" ]; then
-            echo "::error::Missing MOLECULE_STAGING_ADMIN_TOKEN"
+            echo "::error::Missing CP_STAGING_ADMIN_API_TOKEN"
            exit 2
          fi

@ -207,7 +210,7 @@ jobs:
      - name: Teardown safety net
        if: always() && needs.detect-changes.outputs.canvas == 'true'
        env:
-          ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+          ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
        run: |
          set +e
          STATE_FILE=".playwright-staging-state.json"
--- a/.gitea/workflows/e2e-staging-external.yml
+++ b/.gitea/workflows/e2e-staging-external.yml
@ -89,7 +89,10 @@ jobs:

    env:
      MOLECULE_CP_URL: https://staging-api.moleculesai.app
-      MOLECULE_ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+      # 2026-05-11: secret canonicalised from MOLECULE_STAGING_ADMIN_TOKEN
+      # (dead in org secret store) to CP_STAGING_ADMIN_API_TOKEN per
+      # internal#322 — see this PR for the cross-workflow sweep.
+      MOLECULE_ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
      E2E_RUN_ID: "${{ github.run_id }}-${{ github.run_attempt }}"
      E2E_KEEP_ORG: ${{ github.event.inputs.keep_org && '1' || '0' }}
      E2E_STALE_WAIT_SECS: ${{ github.event.inputs.stale_wait_secs || '180' }}
@ -104,7 +107,7 @@ jobs:
            # missing — silent skip would mask infra rot. Manual dispatch
            # gets the same hard-fail; an operator running this on a fork
            # without secrets configured needs to know up-front.
-            echo "::error::MOLECULE_STAGING_ADMIN_TOKEN secret not set (Railway staging CP_ADMIN_API_TOKEN)"
+            echo "::error::CP_STAGING_ADMIN_API_TOKEN secret not set (Railway staging CP_ADMIN_API_TOKEN)"
            exit 2
          fi
          echo "Admin token present ✓"
@ -129,7 +132,7 @@ jobs:
      - name: Teardown safety net (runs on cancel/failure)
        if: always()
        env:
-          ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+          ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
        run: |
          set +e
          orgs=$(curl -sS "$MOLECULE_CP_URL/cp/admin/orgs" \
--- a/.gitea/workflows/e2e-staging-saas.yml
+++ b/.gitea/workflows/e2e-staging-saas.yml
@ -86,7 +86,10 @@ jobs:
      # Single admin-bearer secret drives provision + tenant-token
      # retrieval + teardown. Configure in
      # Settings → Secrets and variables → Actions → Repository secrets.
-      MOLECULE_ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+      # 2026-05-11: secret canonicalised from MOLECULE_STAGING_ADMIN_TOKEN
+      # (dead in org secret store) to CP_STAGING_ADMIN_API_TOKEN per
+      # internal#322 — see this PR for the cross-workflow sweep.
+      MOLECULE_ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
      # MiniMax is the PRIMARY LLM auth path post-2026-05-04. Switched
      # from hermes+OpenAI default after #2578 (the staging OpenAI key
      # account went over quota and stayed dead for 36+ hours, taking
@ -95,7 +98,7 @@ jobs:
      # ANTHROPIC_BASE_URL to api.minimax.io/anthropic and reads
      # MINIMAX_API_KEY at boot — separate billing account so an
      # OpenAI quota collapse no longer wedges the gate. Mirrors the
-      # canary-staging.yml + continuous-synth-e2e.yml migrations.
+      # staging-smoke.yml + continuous-synth-e2e.yml migrations.
      E2E_MINIMAX_API_KEY: ${{ secrets.MOLECULE_STAGING_MINIMAX_API_KEY }}
      # Direct-Anthropic alternative for operators who don't want to
      # set up a MiniMax account (priority below MiniMax — first
@ -122,7 +125,7 @@ jobs:
      - name: Verify admin token present
        run: |
          if [ -z "$MOLECULE_ADMIN_TOKEN" ]; then
-            echo "::error::MOLECULE_STAGING_ADMIN_TOKEN secret not set (Railway staging CP_ADMIN_API_TOKEN)"
+            echo "::error::CP_STAGING_ADMIN_API_TOKEN secret not set (Railway staging CP_ADMIN_API_TOKEN)"
            exit 2
          fi
          echo "Admin token present ✓"
@ -189,7 +192,7 @@ jobs:
      - name: Teardown safety net (runs on cancel/failure)
        if: always()
        env:
-          ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+          ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
        run: |
          # Best-effort: find any e2e-YYYYMMDD-* orgs matching this run and
          # nuke them. Catches the case where the script died before
--- a/.gitea/workflows/e2e-staging-sanity.yml
+++ b/.gitea/workflows/e2e-staging-sanity.yml
@ -11,11 +11,11 @@ name: E2E Staging Sanity (leak-detection self-check)
 #   - `continue-on-error: true` on the job (RFC §1 contract).
 #
 # Periodic assertion that the teardown safety nets in e2e-staging-saas
-# and canary-staging actually work. Runs the E2E harness with
-# E2E_INTENTIONAL_FAILURE=1, which poisons the tenant admin token after
-# the org is provisioned. The workspace-provision step then fails, the
-# script exits non-zero, and the EXIT trap + workflow always()-step
-# must still tear down cleanly.
+# and staging-smoke (formerly canary-staging) actually work. Runs the
+# E2E harness with E2E_INTENTIONAL_FAILURE=1, which poisons the tenant
+# admin token after the org is provisioned. The workspace-provision
+# step then fails, the script exits non-zero, and the EXIT trap +
+# workflow always()-step must still tear down cleanly.

 on:
  schedule:
@ -42,8 +42,11 @@ jobs:

    env:
      MOLECULE_CP_URL: https://staging-api.moleculesai.app
-      MOLECULE_ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
-      E2E_MODE: canary
+      # 2026-05-11: secret canonicalised from MOLECULE_STAGING_ADMIN_TOKEN
+      # (dead in org secret store) to CP_STAGING_ADMIN_API_TOKEN per
+      # internal#322 — see this PR for the cross-workflow sweep.
+      MOLECULE_ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
+      E2E_MODE: smoke
      E2E_RUNTIME: hermes
      E2E_RUN_ID: "sanity-${{ github.run_id }}"
      E2E_INTENTIONAL_FAILURE: "1"
@ -54,7 +57,7 @@ jobs:
      - name: Verify admin token present
        run: |
          if [ -z "$MOLECULE_ADMIN_TOKEN" ]; then
-            echo "::error::MOLECULE_STAGING_ADMIN_TOKEN not set"
+            echo "::error::CP_STAGING_ADMIN_API_TOKEN not set"
            exit 2
          fi

@ -118,7 +121,7 @@ jobs:
      - name: Teardown safety net
        if: always()
        env:
-          ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+          ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
        run: |
          set +e
          orgs=$(curl -sS "$MOLECULE_CP_URL/cp/admin/orgs" \
@ -127,8 +130,14 @@ jobs:
          import json, sys
          d = json.load(sys.stdin)
          today = __import__('datetime').date.today().strftime('%Y%m%d')
+          # Match both the new e2e-smoke- prefix (post-2026-05-11 rename)
+          # and the legacy e2e-canary- prefix for one rollout cycle so
+          # any in-flight org provisioned under the old prefix on an
+          # older runner checkout still gets cleaned up. Remove the
+          # canary fallback after one week of no-old-prefix observations.
+          prefixes = (f'e2e-smoke-{today}-sanity-', f'e2e-canary-{today}-sanity-')
          candidates = [o['slug'] for o in d.get('orgs', [])
-                        if o.get('slug','').startswith(f'e2e-canary-{today}-sanity-')
+                        if any(o.get('slug','').startswith(p) for p in prefixes)
                        and o.get('status') not in ('purged',)]
          print('\n'.join(candidates))
          " 2>/dev/null)
--- a/.gitea/workflows/publish-canvas-image.yml
+++ b/.gitea/workflows/publish-canvas-image.yml
@ -11,7 +11,7 @@ name: publish-canvas-image
 #   - `continue-on-error: true` on each job (RFC §1 contract).
 #   - **Open question for review**: this workflow pushes the canvas
 #     image to `ghcr.io`. GHCR was retired during the 2026-05-06
-#     Gitea migration in favor of ECR (per canary-verify.yml header
+#     Gitea migration in favor of ECR (per staging-verify.yml header
 #     notes). The image may not be consumable post-migration. Two
 #     options for follow-up: (a) retarget to
 #     `153263036946.dkr.ecr.us-east-2.amazonaws.com/molecule-ai/canvas`,
--- a/.gitea/workflows/redeploy-tenants-on-main.yml
+++ b/.gitea/workflows/redeploy-tenants-on-main.yml
@ -32,7 +32,7 @@ name: redeploy-tenants-on-main
 #
 # Registry: ECR (153263036946.dkr.ecr.us-east-2.amazonaws.com/
 # molecule-ai/platform-tenant). GHCR was retired 2026-05-07 during the
-# Gitea suspension migration. The canary-verify.yml promote step now
+# Gitea suspension migration. The staging-verify.yml promote step now
 # uses the same redeploy-fleet endpoint (fixes the silent-GHCR gap).
 #
 # Runtime ordering:
@ -104,7 +104,7 @@ jobs:
        #      `staging-<sha>` to roll back to a known-good build.
        #   2. Default → `staging-<short_head_sha>`. The just-published
        #      digest. Bypasses the `:latest` retag path that's currently
-        #      dead (canary-verify soft-skips without canary fleet, so
+        #      dead (staging-verify soft-skips without canary fleet, so
        #      the only thing retagging `:latest` today is the manual
        #      promote-latest.yml — last run 2026-04-28). Auto-trigger
        #      from workflow_run uses workflow_run.head_sha; manual
@ -359,7 +359,7 @@ jobs:

          # Belt-and-suspenders sanity floor: same logic as the staging
          # variant — see that file's comment for the full rationale.
-          # Floor only applies when fleet >= 4; below that, canary-verify
+          # Floor only applies when fleet >= 4; below that, staging-verify
          # is the actual gate.
          TOTAL_VERIFIED=${#SLUGS[@]}
          if [ $TOTAL_VERIFIED -ge 4 ] && [ $UNREACHABLE_COUNT -gt $((TOTAL_VERIFIED / 2)) ]; then
--- a/.gitea/workflows/redeploy-tenants-on-staging.yml
+++ b/.gitea/workflows/redeploy-tenants-on-staging.yml
@ -21,7 +21,7 @@ name: redeploy-tenants-on-staging
 #
 # Mirror of redeploy-tenants-on-main.yml, with the staging-CP host and
 # the :staging-latest tag. Sister workflow exists for prod (rolls
-# :latest after canary-verify). Both share the same shape — just
+# :latest after staging-verify). Both share the same shape — just
 # different CP_URL + target_tag + admin token secret.
 #
 # Why this workflow exists: publish-workspace-server-image now builds
@ -336,7 +336,7 @@ jobs:
          # crashes on startup), not a teardown race. Hard-fail.
          #
          # Floor only applies when TOTAL_VERIFIED >= 4 — below that, the
-          # canary-verify step is the actual gate for "all tenants down"
+          # staging-verify step is the actual gate for "all tenants down"
          # detection (it runs against the canary first and aborts the
          # rollout if the canary fails to come up). Without the >=4 gate,
          # a 1-tenant fleet (e.g. a single ephemeral e2e-* tenant on a
--- a/.gitea/workflows/canary-staging.yml
+++ b/.gitea/workflows/canary-staging.yml
@ -1,6 +1,8 @@
-name: Canary — staging SaaS smoke (every 30 min)
+name: Staging SaaS smoke (every 30 min)

-# Ported from .github/workflows/canary-staging.yml on 2026-05-11 per RFC
+# Renamed from canary-staging.yml on 2026-05-11 per Hongming directive
+# ("canary naming changed to staging for all"). Originally ported from
+# .github/workflows/canary-staging.yml on 2026-05-11 per RFC
 # internal#219 §1 sweep. Differences from the GitHub version:
 #   - Dropped `workflow_dispatch.inputs` (Gitea 1.22.6 parser rejects them
 #     per feedback_gitea_workflow_dispatch_inputs_unsupported).
@ -21,21 +23,21 @@ name: Canary — staging SaaS smoke (every 30 min)
 # catches drift in the 30-min window between those runs (AMI health, CF
 # cert rotation, WorkOS session stability, etc.).
 #
-# Lean mode: E2E_MODE=canary skips the child workspace + HMA memory +
+# Lean mode: E2E_MODE=smoke skips the child workspace + HMA memory +
 # peers/activity checks. One parent workspace + one A2A turn is enough
 # to signal "SaaS stack end-to-end is alive."

 on:
  schedule:
    # Every 30 min. Cron on GitHub-hosted runners has a known drift of
-    # a few minutes under load — that's fine for a canary.
+    # a few minutes under load — that's fine for a smoke check.
    - cron: '*/30 * * * *'
 # Serialise with the full-SaaS workflow so they don't contend for the
 # same org-create quota on staging. Different group key from
-# e2e-staging-saas since we don't mind queueing canaries behind one
-# full run, but two canaries SHOULD queue against each other.
+# e2e-staging-saas since we don't mind queueing smoke runs behind one
+# full run, but two smoke runs SHOULD queue against each other.
 concurrency:
-  group: canary-staging
+  group: staging-smoke
  cancel-in-progress: false

 permissions:
@ -47,32 +49,47 @@ env:
  GITHUB_SERVER_URL: https://git.moleculesai.app

 jobs:
-  canary:
-    name: Canary smoke
+  smoke:
+    name: Staging SaaS smoke
    runs-on: ubuntu-latest
-    # Phase 3 (RFC #219 §1): surface broken workflows without blocking.
-    continue-on-error: true
+    # NOTE: Phase 3 (RFC #219 §1) `continue-on-error: true` removed
+    # 2026-05-11. The "surface broken workflows without blocking"
+    # rationale was correctly applied to advisory/lint workflows but
+    # wrong for this smoke — it is the 30-min canary cadence for the
+    # entire staging SaaS stack, and silent failure here masks the
+    # exact regressions the smoke exists to surface (AMI rot, CF cert
+    # drift, WorkOS session breakage, secret rotations). Same class of
+    # failure as PR#461 (`sweep-stale-e2e-orgs`) where Phase-3 silent
+    # failure leaked EC2. The four other `e2e-staging-*` workflows
+    # KEEP `continue-on-error: true` per RFC #219 §1 — they are
+    # advisory and matrix-style; this one is the canary. A follow-up
+    # `notify-failure` step below also surfaces breakage to ops even
+    # if branch-protection wiring is adjusted to keep this off the
+    # required-checks list.
    # 25 min headroom over the 15-min TLS-readiness deadline in
    # tests/e2e/test_staging_full_saas.sh (#2107). Without the buffer
    # the job is killed at the wall-clock 15:00 mark BEFORE the bash
    # `fail` + diagnostic burst can fire, leaving every cancellation
-    # silent. Sibling staging E2E jobs run at 20-45 min — keeping
-    # canary tighter than them so a true wedge still surfaces here
+    # silent. Sibling staging E2E jobs run at 20-45 min — keeping the
+    # smoke tighter than them so a true wedge still surfaces here
    # first.
    timeout-minutes: 25

    env:
      MOLECULE_CP_URL: https://staging-api.moleculesai.app
-      MOLECULE_ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
-      # MiniMax is the canary's PRIMARY LLM auth path post-2026-05-04.
+      # 2026-05-11: secret canonicalised from MOLECULE_STAGING_ADMIN_TOKEN
+      # (dead in org secret store) to CP_STAGING_ADMIN_API_TOKEN per
+      # internal#322 — see this PR for the cross-workflow sweep.
+      MOLECULE_ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
+      # MiniMax is the smoke's PRIMARY LLM auth path post-2026-05-04.
      # Switched from hermes+OpenAI after #2578 (the staging OpenAI key
      # account went over quota and stayed dead for 36+ hours, taking
-      # the canary red the entire time). claude-code template's
+      # the smoke red the entire time). claude-code template's
      # `minimax` provider routes ANTHROPIC_BASE_URL to
      # api.minimax.io/anthropic and reads MINIMAX_API_KEY at boot —
      # ~5-10x cheaper per token than gpt-4.1-mini AND on a separate
      # billing account, so OpenAI quota collapse no longer wedges the
-      # canary. Mirrors the migration continuous-synth-e2e.yml made on
+      # smoke. Mirrors the migration continuous-synth-e2e.yml made on
      # 2026-05-03 (#265) for the same reason. tests/e2e/test_staging_
      # full_saas.sh branches SECRETS_JSON on which key is present —
      # MiniMax wins when set.
@ -86,16 +103,16 @@ jobs:
      # E2E_RUNTIME=hermes overridden via workflow_dispatch can still
      # exercise the OpenAI path without re-editing the workflow.
      E2E_OPENAI_API_KEY: ${{ secrets.MOLECULE_STAGING_OPENAI_API_KEY }}
-      E2E_MODE: canary
+      E2E_MODE: smoke
      E2E_RUNTIME: claude-code
-      # Pin the canary to a specific MiniMax model rather than relying
+      # Pin the smoke to a specific MiniMax model rather than relying
      # on the per-runtime default (which could resolve to "sonnet" →
      # direct Anthropic and defeat the cost saving). M2.7-highspeed
      # is "Token Plan only" but cheap-per-token and fast.
      E2E_MODEL_SLUG: MiniMax-M2.7-highspeed
-      E2E_RUN_ID: "canary-${{ github.run_id }}"
+      E2E_RUN_ID: "smoke-${{ github.run_id }}"
      # Debug-only: when an operator dispatches with keep_on_failure=true,
-      # the canary script's E2E_KEEP_ORG=1 path skips teardown so the
+      # the smoke script's E2E_KEEP_ORG=1 path skips teardown so the
      # tenant org + EC2 stay alive for SSM-based log capture. Cron runs
      # never set this (the input only exists on workflow_dispatch) so
      # unattended cron always tears down. See molecule-core#129
@ -109,7 +126,7 @@ jobs:
      - name: Verify admin token present
        run: |
          if [ -z "$MOLECULE_ADMIN_TOKEN" ]; then
-            echo "::error::MOLECULE_STAGING_ADMIN_TOKEN not set"
+            echo "::error::CP_STAGING_ADMIN_API_TOKEN not set"
            exit 2
          fi

@ -119,7 +136,7 @@ jobs:
          # langgraph (operator-dispatched only) use OpenAI. Hard-fail
          # rather than soft-skip per the lesson from synth E2E #2578:
          # an empty key silently falls through to the wrong
-          # SECRETS_JSON branch and the canary fails 5 min later with
+          # SECRETS_JSON branch and the smoke fails 5 min later with
          # a confusing auth error instead of the clean "secret
          # missing" message at the top.
          case "${E2E_RUNTIME}" in
@ -155,8 +172,8 @@ jobs:
          fi
          echo "LLM key present ✓ (runtime=${E2E_RUNTIME}, key=${required_secret_name}, len=${#required_secret_value})"

-      - name: Canary run
-        id: canary
+      - name: Smoke run
+        id: smoke
        run: bash tests/e2e/test_staging_full_saas.sh

      # Alerting: open a sticky issue on the FIRST failure; comment on
@ -184,6 +201,9 @@ jobs:
        run: |
          set -euo pipefail
          API="${SERVER_URL%/}/api/v1"
+          # Title kept stable across the canary-staging.yml → staging-smoke.yml
+          # rename (2026-05-11) so any open alert issue from the old name
+          # still title-matches and auto-closes on the next green run.
          TITLE="Canary failing: staging SaaS smoke"
          RUN_URL="${SERVER_URL}/${REPO}/actions/runs/${RUN_ID}"

@ -194,18 +214,18 @@ jobs:
          if [ -n "$EXISTING" ]; then
            curl -fsS -X POST -H "Authorization: token $GITEA_TOKEN" -H "Content-Type: application/json" \
              "${API}/repos/${REPO}/issues/${EXISTING}/comments" \
-              -d "$(jq -nc --arg run "$RUN_URL" '{body: ("Canary still failing. " + $run)}')" >/dev/null
+              -d "$(jq -nc --arg run "$RUN_URL" '{body: ("Smoke still failing. " + $run)}')" >/dev/null
            echo "Commented on existing issue #${EXISTING}"
          else
            NOW=$(date -u +%Y-%m-%dT%H:%M:%SZ)
            BODY=$(jq -nc --arg t "$TITLE" --arg now "$NOW" --arg run "$RUN_URL" \
-              '{title: $t, body: ("Canary run failed at " + $now + ".\n\nRun: " + $run + "\n\nThis issue auto-closes on the next green canary run. Consecutive failures add a comment here rather than a new issue.")}')
+              '{title: $t, body: ("Smoke run failed at " + $now + ".\n\nRun: " + $run + "\n\nThis issue auto-closes on the next green smoke run. Consecutive failures add a comment here rather than a new issue.")}')
            curl -fsS -X POST -H "Authorization: token $GITEA_TOKEN" -H "Content-Type: application/json" \
              "${API}/repos/${REPO}/issues" -d "$BODY" >/dev/null
-            echo "Opened canary failure issue (first red)"
+            echo "Opened smoke failure issue (first red)"
          fi

-      - name: Auto-close canary issue on success (Gitea API)
+      - name: Auto-close smoke issue on success (Gitea API)
        if: success()
        env:
          GITEA_TOKEN: ${{ secrets.GITHUB_TOKEN }}
@ -215,6 +235,8 @@ jobs:
        run: |
          set -euo pipefail
          API="${SERVER_URL%/}/api/v1"
+          # Title kept stable across the canary-staging.yml → staging-smoke.yml
+          # rename so open alert issues from the old name still match.
          TITLE="Canary failing: staging SaaS smoke"

          NUMS=$(curl -fsS -H "Authorization: token $GITEA_TOKEN" \
@ -225,37 +247,36 @@ jobs:
          for N in $NUMS; do
            curl -fsS -X POST -H "Authorization: token $GITEA_TOKEN" -H "Content-Type: application/json" \
              "${API}/repos/${REPO}/issues/${N}/comments" \
-              -d "$(jq -nc --arg now "$NOW" '{body: ("Canary recovered at " + $now + ". Closing.")}')" >/dev/null
+              -d "$(jq -nc --arg now "$NOW" '{body: ("Smoke recovered at " + $now + ". Closing.")}')" >/dev/null
            curl -fsS -X PATCH -H "Authorization: token $GITEA_TOKEN" -H "Content-Type: application/json" \
              "${API}/repos/${REPO}/issues/${N}" -d '{"state":"closed"}' >/dev/null
-            echo "Closed recovered canary issue #${N}"
+            echo "Closed recovered smoke issue #${N}"
          done

      - name: Teardown safety net
        if: always()
        env:
-          ADMIN_TOKEN: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKEN }}
+          ADMIN_TOKEN: ${{ secrets.CP_STAGING_ADMIN_API_TOKEN }}
        run: |
          set +e
          # Slug prefix matches what test_staging_full_saas.sh emits
-          # in canary mode:
-          #   SLUG="e2e-canary-$(date +%Y%m%d)-${RUN_ID_SUFFIX}"
-          # Earlier this was `e2e-{today}-canary-` — that was the
-          # full-mode pattern (date FIRST, mode SECOND); canary slugs
-          # have mode FIRST, date SECOND. The mismatch silently
-          # never matched, leaving every cancelled-canary EC2 alive
-          # until the once-an-hour sweep eventually caught it
-          # (incident 2026-04-26 21:03Z: 1h25m EC2 leak before manual
-          # cleanup; same gap on three earlier cancellations today).
+          # in smoke mode:
+          #   SLUG="e2e-smoke-$(date +%Y%m%d)-${RUN_ID_SUFFIX}"
+          # Earlier (pre-2026-05-11 canary→staging rename) the prefix was
+          # `e2e-canary-`; both prefixes are matched here for one
+          # release cycle so cleanup still catches any in-flight org
+          # provisioned under the old prefix on an older runner that
+          # hasn't picked up the renamed script. Remove the canary
+          # fallback after one week of no-old-prefix observations.
          orgs=$(curl -sS "$MOLECULE_CP_URL/cp/admin/orgs" \
            -H "Authorization: Bearer $ADMIN_TOKEN" 2>/dev/null \
            | python3 -c "
          import json, sys, os, datetime
          run_id = os.environ.get('GITHUB_RUN_ID', '')
          d = json.load(sys.stdin)
-          # Scope to slugs from THIS canary run when GITHUB_RUN_ID is
-          # available; the canary workflow sets E2E_RUN_ID='canary-\${run_id}'
-          # so the slug suffix is '-canary-\${run_id}-...'. Mirrors the
+          # Scope to slugs from THIS smoke run when GITHUB_RUN_ID is
+          # available; the smoke workflow sets E2E_RUN_ID='smoke-\${run_id}'
+          # so the slug suffix is '-smoke-\${run_id}-...'. Mirrors the
          # full-mode safety net's per-run scoping (e2e-staging-saas.yml)
          # added after the 2026-04-21 cross-run cleanup incident.
          # Sweep both today AND yesterday's UTC dates so a run that
@ -265,9 +286,11 @@ jobs:
          yesterday = today - datetime.timedelta(days=1)
          dates = (today.strftime('%Y%m%d'), yesterday.strftime('%Y%m%d'))
          if run_id:
-              prefixes = tuple(f'e2e-canary-{d}-canary-{run_id}' for d in dates)
+              prefixes = tuple(f'e2e-smoke-{d}-smoke-{run_id}' for d in dates) \
+                       + tuple(f'e2e-canary-{d}-canary-{run_id}' for d in dates)
          else:
-              prefixes = tuple(f'e2e-canary-{d}-' for d in dates)
+              prefixes = tuple(f'e2e-smoke-{d}-' for d in dates) \
+                       + tuple(f'e2e-canary-{d}-' for d in dates)
          candidates = [o['slug'] for o in d.get('orgs', [])
                        if any(o.get('slug','').startswith(p) for p in prefixes)
                        and o.get('status') not in ('purged',)]
@ -280,8 +303,8 @@ jobs:
          # stale sweep caught it (up to 2h later). Now we capture the
          # response code and surface non-2xx as a workflow warning, so
          # the run page shows which slug leaked. We still don't `exit 1`
-          # on cleanup failure — a single-canary cleanup miss shouldn't
-          # fail-flag the canary itself when the actual smoke check
+          # on cleanup failure — a single-smoke cleanup miss shouldn't
+          # fail-flag the smoke itself when the actual smoke check
          # passed. The sweep-stale-e2e-orgs cron (now every 15 min,
          # 30-min threshold) is the safety net for whatever slips past.
          # See molecule-controlplane#420.
@ -290,21 +313,34 @@ jobs:
            # Tempfile-routed -w + set +e/-e prevents curl-exit-code
            # pollution of the captured status (lint-curl-status-capture.yml).
            set +e
-            curl -sS -o /tmp/canary-cleanup.out -w "%{http_code}" \
+            curl -sS -o /tmp/smoke-cleanup.out -w "%{http_code}" \
              -X DELETE "$MOLECULE_CP_URL/cp/admin/tenants/$slug" \
              -H "Authorization: Bearer $ADMIN_TOKEN" \
              -H "Content-Type: application/json" \
-              -d "{\"confirm\":\"$slug\"}" >/tmp/canary-cleanup.code
+              -d "{\"confirm\":\"$slug\"}" >/tmp/smoke-cleanup.code
            set -e
-            code=$(cat /tmp/canary-cleanup.code 2>/dev/null || echo "000")
+            code=$(cat /tmp/smoke-cleanup.code 2>/dev/null || echo "000")
            if [ "$code" = "200" ] || [ "$code" = "204" ]; then
              echo "[teardown] deleted $slug (HTTP $code)"
            else
-              echo "::warning::canary teardown for $slug returned HTTP $code — sweep-stale-e2e-orgs will catch it within ~45 min. Body: $(head -c 300 /tmp/canary-cleanup.out 2>/dev/null)"
+              echo "::warning::smoke teardown for $slug returned HTTP $code — sweep-stale-e2e-orgs will catch it within ~45 min. Body: $(head -c 300 /tmp/smoke-cleanup.out 2>/dev/null)"
              leaks+=("$slug")
            fi
          done
          if [ ${#leaks[@]} -gt 0 ]; then
-            echo "::warning::canary teardown left ${#leaks[@]} leak(s): ${leaks[*]}"
+            echo "::warning::smoke teardown left ${#leaks[@]} leak(s): ${leaks[*]}"
          fi
          exit 0
+
+      - name: Notify on smoke failure
+        # Fail-loud companion to dropping `continue-on-error: true`.
+        # The Open-issue-on-failure step above handles the human-facing
+        # alert; this step emits a clearly-tagged ::error:: line that
+        # log-tail consumers (Loki SOPRefireRule, orchestrator triage
+        # loop) can grep on. Mirrors PR#461's sweep-stale-e2e-orgs
+        # pattern. Runs AFTER the teardown safety net (which is
+        # if: always()) so failures don't suppress cleanup.
+        if: failure()
+        run: |
+          echo "::error::staging-smoke FAILED — staging SaaS canary is red. See prior step logs + the auto-filed alert issue. Common causes: (a) CP_STAGING_ADMIN_API_TOKEN secret missing/rotated, (b) staging-api.moleculesai.app 5xx, (c) MiniMax/Anthropic LLM key dead, (d) AMI/CF/WorkOS drift. The 30-min cron will retry, but a chronic red here indicates the staging SaaS stack is broken end-to-end."
+          exit 1
--- a/.gitea/workflows/staging-verify.yml
+++ b/.gitea/workflows/staging-verify.yml
@ -1,6 +1,8 @@
-name: canary-verify
+name: Staging verify

-# Ported from .github/workflows/canary-verify.yml on 2026-05-11 per RFC
+# Renamed from canary-verify.yml on 2026-05-11 per Hongming directive
+# ("canary naming changed to staging for all"). Originally ported from
+# .github/workflows/canary-verify.yml on 2026-05-11 per RFC
 # internal#219 §1 sweep. Differences from the GitHub version:
 #   - Dropped `workflow_dispatch.inputs` (Gitea 1.22.6 parser rejects them
 #     per feedback_gitea_workflow_dispatch_inputs_unsupported).
@ -23,13 +25,22 @@ name: canary-verify
 # digest. On red, :latest stays on the prior known-good digest and
 # prod is untouched.
 #
+# Terminology note (2026-05-11): The deployment STRATEGY here is still
+# called "canary release" (a small subset of tenants gets the new image
+# first, the rest follow on green). The "canary" word stays for the
+# pre-fan-out cohort concept (see docs/architecture/canary-release.md
+# and CANARY_SLUG in redeploy-tenants-on-*.yml). What changed is the
+# FILE NAME and the SECRETS feeding this workflow — both are renamed
+# to drop the redundant "canary-" prefix that conflated workflow
+# identity with deployment strategy.
+#
 # Registry note (2026-05-10): This workflow previously used GHCR
 # (ghcr.io/molecule-ai/platform-tenant) — that registry was retired
 # during the 2026-05-06 Gitea suspension migration when publish-
 # workspace-server-image.yml switched to the operator's ECR org
 # (153263036946.dkr.ecr.us-east-2.amazonaws.com/molecule-ai/
 # platform-tenant). The GHCR → ECR migration was never applied to
-# this file, so canary-verify was silently smoke-testing the stale
+# this file, so this workflow was silently smoke-testing the stale
 # GHCR image while the actual staging/prod tenants ran the ECR image.
 # Result: smoke tests could not catch a broken ECR build. Fix:
 #   - Wait step: reads SHA from running canary /health (tenant-
@ -43,8 +54,9 @@ name: canary-verify
 #     to ECR on staging and main merges.
 #   - Canary tenants are configured to pull :staging-<sha> from ECR
 #     (TENANT_IMAGE env set to the ECR :staging-<sha> tag).
-#   - Repo secrets CANARY_TENANT_URLS / CANARY_ADMIN_TOKENS /
-#     CANARY_CP_SHARED_SECRET are populated.
+#   - Repo secrets MOLECULE_STAGING_TENANT_URLS /
+#     MOLECULE_STAGING_ADMIN_TOKENS / MOLECULE_STAGING_CP_SHARED_SECRET
+#     are populated.

 on:
  workflow_run:
@ -65,7 +77,7 @@ env:
  GITHUB_SERVER_URL: https://git.moleculesai.app

 jobs:
-  canary-smoke:
+  staging-smoke:
    # Skip when the upstream workflow failed — no image to test against.
    # workflow_dispatch trigger dropped in this Gitea port; only the
    # workflow_run path remains.
@ -97,15 +109,15 @@ jobs:
        # other registry — the canary is telling us what it's actually
        # running, which is the ground truth for smoke testing.
        env:
-          CANARY_TENANT_URLS: ${{ secrets.CANARY_TENANT_URLS }}
+          MOLECULE_STAGING_TENANT_URLS: ${{ secrets.MOLECULE_STAGING_TENANT_URLS }}
          EXPECTED_SHA: ${{ steps.compute.outputs.sha }}
        run: |
-          if [ -z "$CANARY_TENANT_URLS" ]; then
+          if [ -z "$MOLECULE_STAGING_TENANT_URLS" ]; then
            echo "No canary URLs configured — falling back to 60s wait"
            sleep 60
            exit 0
          fi
-          IFS=',' read -ra URLS <<< "$CANARY_TENANT_URLS"
+          IFS=',' read -ra URLS <<< "$MOLECULE_STAGING_TENANT_URLS"
          MAX_WAIT=420  # 7 minutes
          INTERVAL=30
          ELAPSED=0
@ -129,7 +141,7 @@ jobs:
          done
          echo "Timeout after ${MAX_WAIT}s — proceeding anyway (smoke suite will validate)"

-      - name: Run canary smoke suite
+      - name: Run staging smoke suite
        id: smoke
        # Graceful-skip when no canary fleet is configured (Phase 2 not yet
        # stood up — see molecule-controlplane/docs/canary-tenants.md).
@ -138,29 +150,29 @@ jobs:
        # promote-latest.yml is the release gate while canary is absent.
        # Once the fleet is real: delete the early-exit branch.
        env:
-          CANARY_TENANT_URLS: ${{ secrets.CANARY_TENANT_URLS }}
-          CANARY_ADMIN_TOKENS: ${{ secrets.CANARY_ADMIN_TOKENS }}
-          CANARY_CP_BASE_URL: https://staging-api.moleculesai.app
-          CANARY_CP_SHARED_SECRET: ${{ secrets.CANARY_CP_SHARED_SECRET }}
+          MOLECULE_STAGING_TENANT_URLS: ${{ secrets.MOLECULE_STAGING_TENANT_URLS }}
+          MOLECULE_STAGING_ADMIN_TOKENS: ${{ secrets.MOLECULE_STAGING_ADMIN_TOKENS }}
+          MOLECULE_STAGING_CP_BASE_URL: https://staging-api.moleculesai.app
+          MOLECULE_STAGING_CP_SHARED_SECRET: ${{ secrets.MOLECULE_STAGING_CP_SHARED_SECRET }}
        run: |
          set -euo pipefail
-          if [ -z "${CANARY_TENANT_URLS:-}" ] \
-            || [ -z "${CANARY_ADMIN_TOKENS:-}" ] \
-            || [ -z "${CANARY_CP_SHARED_SECRET:-}" ]; then
+          if [ -z "${MOLECULE_STAGING_TENANT_URLS:-}" ] \
+            || [ -z "${MOLECULE_STAGING_ADMIN_TOKENS:-}" ] \
+            || [ -z "${MOLECULE_STAGING_CP_SHARED_SECRET:-}" ]; then
            {
-              echo "## ⚠️ canary-verify skipped"
+              echo "## ⚠️ staging-verify skipped"
              echo
-              echo "One or more canary secrets are unset (\`CANARY_TENANT_URLS\`, \`CANARY_ADMIN_TOKENS\`, \`CANARY_CP_SHARED_SECRET\`)."
+              echo "One or more canary secrets are unset (\`MOLECULE_STAGING_TENANT_URLS\`, \`MOLECULE_STAGING_ADMIN_TOKENS\`, \`MOLECULE_STAGING_CP_SHARED_SECRET\`)."
              echo "Phase 2 canary fleet has not been stood up yet —"
              echo "see [canary-tenants.md](https://git.moleculesai.app/molecule-ai/molecule-controlplane/blob/main/docs/canary-tenants.md)."
              echo
              echo "**Skipped — promote-to-latest will NOT auto-fire.** Dispatch \`promote-latest.yml\` manually when ready."
            } >> "$GITHUB_STEP_SUMMARY"
            echo "ran=false" >> "$GITHUB_OUTPUT"
-            echo "::notice::canary-verify: skipped — no canary fleet configured"
+            echo "::notice::staging-verify: skipped — no canary fleet configured"
            exit 0
          fi
-          bash scripts/canary-smoke.sh
+          bash scripts/staging-smoke.sh
          echo "ran=true" >> "$GITHUB_OUTPUT"

      - name: Summary on failure
@ -173,7 +185,7 @@ jobs:
            echo ":latest stays pinned to the prior good digest — prod is untouched."
            echo
            echo "Fix forward and merge again, or investigate the specific failed"
-            echo "assertions in the canary-smoke step log above."
+            echo "assertions in the staging-smoke step log above."
          } >> "$GITHUB_STEP_SUMMARY"

  promote-to-latest:
@ -188,13 +200,13 @@ jobs:
    # silently promoting a stale GHCR image while actual prod tenants
    # pulled from ECR. Canary smoke tests were GHCR-targeted and could
    # not catch a broken ECR build.
-    needs: canary-smoke
-    if: ${{ needs.canary-smoke.result == 'success' && needs.canary-smoke.outputs.smoke_ran == 'true' }}
+    needs: staging-smoke
+    if: ${{ needs.staging-smoke.result == 'success' && needs.staging-smoke.outputs.smoke_ran == 'true' }}
    runs-on: ubuntu-latest
    # Phase 3 (RFC #219 §1): surface broken workflows without blocking.
    continue-on-error: true
    env:
-      SHA: ${{ needs.canary-smoke.outputs.sha }}
+      SHA: ${{ needs.staging-smoke.outputs.sha }}
      CP_URL: ${{ vars.CP_URL || 'https://staging-api.moleculesai.app' }}
      # CP_ADMIN_API_TOKEN gates write access to the redeploy endpoint.
      # Stored at the repo level so all workflows pick it up automatically.
@ -264,9 +276,9 @@ jobs:
      - name: Summary
        run: |
          {
-            echo "## Canary verified — :latest promoted via CP redeploy-fleet"
+            echo "## Staging verified — :latest promoted via CP redeploy-fleet"
            echo ""
-            echo "- **Target tag:** \`staging-${{ needs.canary-smoke.outputs.sha }}\`"
+            echo "- **Target tag:** \`staging-${{ needs.staging-smoke.outputs.sha }}\`"
            echo "- **Registry:** ECR (\`${TENANT_IMAGE_NAME}\`)"
            echo "- **Canary slug:** \`${CANARY_SLUG:-<none>}\` (soak ${SOAK_SECONDS}s)"
            echo "- **Batch size:** ${BATCH_SIZE:-3}"
--- a/.gitea/workflows/sweep-stale-e2e-orgs.yml
+++ b/.gitea/workflows/sweep-stale-e2e-orgs.yml
@ -99,7 +99,8 @@ jobs:

          # Filter:
          #   1. slug starts with one of the ephemeral test prefixes:
-          #        - 'e2e-'    — covers e2e-canary-, e2e-canvas-*, etc.
+          #        - 'e2e-'    — covers e2e-smoke- (formerly e2e-canary-),
+          #                      e2e-canvas-*, etc.
          #        - 'rt-e2e-' — runtime-test harness fixtures (RFC #2251);
          #                      missing this prefix left two such tenants
          #                      orphaned 8h on staging (2026-05-03), then
--- a/docs/architecture/canary-release.md
+++ b/docs/architecture/canary-release.md
@ -2,7 +2,7 @@

 How a workspace-server code change reaches the prod tenant fleet — and how to stop it if something's wrong.

-> **⚠️ State note (2026-04-22):** this doc describes the **intended design**. As of this write, the canary fleet described below is **not actually running** — no canary tenants are provisioned, `CANARY_TENANT_URLS` / `CANARY_ADMIN_TOKENS` / `CANARY_CP_SHARED_SECRET` are empty in repo secrets, and `canary-verify.yml` fails every run.
+> **⚠️ State note (2026-04-22, secret names refreshed 2026-05-11):** this doc describes the **intended design**. As of this write, the canary fleet described below is **not actually running** — no canary tenants are provisioned, `MOLECULE_STAGING_TENANT_URLS` / `MOLECULE_STAGING_ADMIN_TOKENS` / `MOLECULE_STAGING_CP_SHARED_SECRET` are empty in repo secrets, and `staging-verify.yml` (formerly `canary-verify.yml`) fails every run.
 >
 > Current merges gate on manual `promote-latest.yml` dispatches, not canary. See [molecule-controlplane/docs/canary-tenants.md](https://git.moleculesai.app/molecule-ai/molecule-controlplane/src/branch/main/docs/canary-tenants.md) for the Phase 1 code work that's already shipped + the Phase 2 plan for actually standing up the fleet + a "should we even do this now?" decision framework.
 >
@ -22,7 +22,7 @@ publish-workspace-server-image.yml   ← pushes :staging-<sha> ONLY
 Canary tenants auto-update to :staging-<sha>
      │   (5-min auto-updater cycle on each canary EC2)
      ▼
-canary-verify.yml waits 6 min, runs scripts/canary-smoke.sh
+staging-verify.yml waits 6 min, runs scripts/staging-smoke.sh
      │
      ├─► GREEN → crane tag :staging-<sha> → :latest
      │                                       │
@ -42,7 +42,7 @@ Canary tenants are configured to pull `:staging-<sha>` (not `:latest`) via `TENA

 ## Smoke suite

-`scripts/canary-smoke.sh` hits each canary tenant (URL + ADMIN_TOKEN pair) and asserts:
+`scripts/staging-smoke.sh` hits each canary tenant (URL + ADMIN_TOKEN pair) and asserts:

 - `/admin/liveness` returns a subsystems map (tenant booted, AdminAuth reachable)
 - `/workspaces` returns a JSON array (wsAuth + DB healthy)
@ -59,8 +59,8 @@ Expand by editing the script — each `check "name" "expected" "$response"` call
 3. Re-trigger provision (or delete + recreate if the org was already provisioned into staging) — the fresh EC2 lands in the canary AWS account (see internal runbook for the specific ID)

 Then set repo secrets:
- `CANARY_TENANT_URLS` — append the new tenant's URL
- `CANARY_ADMIN_TOKENS` — append its ADMIN_TOKEN in the same position
+- `MOLECULE_STAGING_TENANT_URLS` — append the new tenant's URL
+- `MOLECULE_STAGING_ADMIN_TOKENS` — append its ADMIN_TOKEN in the same position

 ## Rolling back `:latest`

--- a/runbooks/gitea-actions-migration-checklist.md
+++ b/runbooks/gitea-actions-migration-checklist.md
@ -50,7 +50,7 @@ pipeline.
 | `check-merge-group-trigger.yml` | The workflow's own header (lines 18-23) documents that it's vacuously satisfied on Gitea — Gitea has no merge queue, no `merge_group:` event type, no `gh-readonly-queue/...` refs. Nothing to lint. |
 | `codeql.yml` | The workflow's own header (lines 3-67) documents that `github/codeql-action/init@v4` hits api.github.com bundle endpoints not implemented by Gitea (observed: `::error::404 page not found` in Initialize CodeQL step). Per Hongming decision 2026-05-07 (task #156): CodeQL is ADVISORY/non-blocking until a Gitea-compatible SAST pipeline lands. Replacement options (Semgrep self-host, Sonatype, GitHub-mirror-for-SAST) tracked in #156. |
 | `pr-guards.yml` | The workflow's own header documents that Gitea has no `gh pr merge --auto` primitive — the guard is a structural no-op on Gitea. Branch protection on `main` does NOT reference any `pr-guards` check name; deletion is safe. |
-| `promote-latest.yml` | Uses `imjasonh/setup-crane` against `ghcr.io/molecule-ai/platform` — the GHCR registry was retired during the 2026-05-06 Gitea migration (per `canary-verify.yml` header notes, the canonical tenant image moved to ECR `153263036946.dkr.ecr.us-east-2.amazonaws.com/molecule-ai/platform-tenant`). The workflow can no longer find any image to retag. Follow-up issue suggested if an ECR-based retag promote is desired. |
+| `promote-latest.yml` | Uses `imjasonh/setup-crane` against `ghcr.io/molecule-ai/platform` — the GHCR registry was retired during the 2026-05-06 Gitea migration (per `staging-verify.yml` header notes — file was renamed from `canary-verify.yml` on 2026-05-11; the canonical tenant image moved to ECR `153263036946.dkr.ecr.us-east-2.amazonaws.com/molecule-ai/platform-tenant`). The workflow can no longer find any image to retag. Follow-up issue suggested if an ECR-based retag promote is desired. |

 ## Category C — ported to .gitea/

--- a/scripts/README.md
+++ b/scripts/README.md
@ -43,7 +43,7 @@ endpoint handler for the supported range.
 - `cleanup-rogue-workspaces.sh` — emergency teardown for leaked
  workspaces. Prompts for confirmation. Pair with the harnesses if a
  cleanup trap fails (see `cleanup_*_failed` events).
- `canary-smoke.sh` — quick smoke test for canary releases.
+- `staging-smoke.sh` — quick smoke test for the staging canary fleet (formerly `canary-smoke.sh`).
 - `dev-start.sh` — local-dev platform bring-up.

 The rest are self-documenting in their header comments.
--- a/scripts/staging-smoke.sh
+++ b/scripts/staging-smoke.sh
@ -1,29 +1,40 @@
 #!/bin/bash
-# canary-smoke.sh — runs the post-deploy smoke suite against the
-# staging canary tenant fleet. Called by the canary-verify.yml GitHub
+# staging-smoke.sh — runs the post-deploy smoke suite against the
+# staging canary tenant fleet. Called by the staging-verify.yml Gitea
 # Actions workflow after a new workspace-server image lands in ECR;
 # exits non-zero on any failure so the workflow can block the
 # redeploy-fleet promotion that would otherwise release broken code
 # to the prod tenant fleet.
 #
+# Naming note (2026-05-11): The script (and its input env vars) were
+# renamed from canary-smoke.sh / CANARY_* to staging-smoke.sh /
+# MOLECULE_STAGING_* per Hongming directive. The tested COHORT is still
+# called the "canary fleet" (a small subset of staging tenants that
+# ingest :staging-<sha> before the rest of the fleet); that strategy
+# concept is unchanged.
+#
 # Registry note: GHCR was retired 2026-05-06. Images are now pushed
 # to the operator's ECR org (153263036946.dkr.ecr.us-east-2.amazonaws.com/
 # molecule-ai/platform-tenant). The registry URL is a runtime concern for
 # the CI push step; this script tests the running tenant directly.
 #
 # Environment:
-#   CANARY_TENANT_URLS       space-sep list of canary tenant base URLs
-#                            (e.g. "https://canary-pm.staging.moleculesai.app
-#                                   https://canary-mcp.staging.moleculesai.app")
-#   CANARY_ADMIN_TOKENS      space-sep list of ADMIN_TOKENs, positionally
-#                            matched to CANARY_TENANT_URLS. Canary tenants
-#                            are provisioned with known ADMIN_TOKENs so CI
-#                            can hit their admin-gated endpoints.
-#   CANARY_CP_BASE_URL       CP base URL the canaries call back to
-#                            (https://staging-api.moleculesai.app)
-#   CANARY_CP_SHARED_SECRET  matches CP's PROVISION_SHARED_SECRET so this
-#                            script can also exercise /cp/workspaces/* via
-#                            the canary's own CPProvisioner identity.
+#   MOLECULE_STAGING_TENANT_URLS       space-sep list of canary tenant base
+#                                       URLs (e.g. "https://canary-pm.staging.
+#                                       moleculesai.app https://canary-mcp.
+#                                       staging.moleculesai.app")
+#   MOLECULE_STAGING_ADMIN_TOKENS      space-sep list of ADMIN_TOKENs,
+#                                       positionally matched to
+#                                       MOLECULE_STAGING_TENANT_URLS.
+#                                       Canary tenants are provisioned with
+#                                       known ADMIN_TOKENs so CI can hit
+#                                       their admin-gated endpoints.
+#   MOLECULE_STAGING_CP_BASE_URL       CP base URL the canaries call back to
+#                                       (https://staging-api.moleculesai.app)
+#   MOLECULE_STAGING_CP_SHARED_SECRET  matches CP's PROVISION_SHARED_SECRET
+#                                       so this script can also exercise
+#                                       /cp/workspaces/* via the canary's
+#                                       own CPProvisioner identity.
 #
 # Exit codes: 0 = all green, 1 = assertion failure, 2 = setup/env problem.

@ -31,12 +42,12 @@ set -euo pipefail

 # ── Setup ────────────────────────────────────────────────────────────────

-: "${CANARY_TENANT_URLS:?space-sep list of canary base URLs required}"
-: "${CANARY_ADMIN_TOKENS:?space-sep list of ADMIN_TOKENs required, same order as URLs}"
-: "${CANARY_CP_BASE_URL:?CP base URL required}"
+: "${MOLECULE_STAGING_TENANT_URLS:?space-sep list of canary base URLs required}"
+: "${MOLECULE_STAGING_ADMIN_TOKENS:?space-sep list of ADMIN_TOKENs required, same order as URLs}"
+: "${MOLECULE_STAGING_CP_BASE_URL:?CP base URL required}"

-read -r -a URLS <<< "$CANARY_TENANT_URLS"
-read -r -a TOKENS <<< "$CANARY_ADMIN_TOKENS"
+read -r -a URLS <<< "$MOLECULE_STAGING_TENANT_URLS"
+read -r -a TOKENS <<< "$MOLECULE_STAGING_ADMIN_TOKENS"

 if [ "${#URLS[@]}" -ne "${#TOKENS[@]}" ]; then
  echo "ERROR: URLS(${#URLS[@]}) and TOKENS(${#TOKENS[@]}) length mismatch" >&2
@ -69,7 +80,7 @@ check() {
 # tenant never gets the wrong token.
 acurl() {
  local base="$1" token="$2"; shift 2
-  curl -sS --max-time 20 -H "Authorization: Bearer $token" "$@" -- "$base${CANARY_ACURL_PATH:-}"
+  curl -sS --max-time 20 -H "Authorization: Bearer $token" "$@" -- "$base${ACURL_PATH:-}"
 }

 # ── Checks (run per canary tenant) ───────────────────────────────────────
@ -80,7 +91,7 @@ for i in "${!URLS[@]}"; do
  printf "\n── %s ──\n" "$base"

  # 1. Liveness — the tenant is up and responding to admin auth.
-  CANARY_ACURL_PATH="/admin/liveness" resp=$(acurl "$base" "$token" || true)
+  ACURL_PATH="/admin/liveness" resp=$(acurl "$base" "$token" || true)
  check "liveness returns a subsystems map" '"subsystems"' "$resp"

  # 2. CP env refresh — the workspace-server fetched MOLECULE_CP_SHARED_SECRET
@ -89,25 +100,25 @@ for i in "${!URLS[@]}"; do
  # booted without crashing on the refresh call. A startup failure in
  # refreshEnvFromCP logs but still boots (best-effort semantics), so
  # this is a sanity check, not a proof.
-  CANARY_ACURL_PATH="/workspaces" resp=$(acurl "$base" "$token" || true)
+  ACURL_PATH="/workspaces" resp=$(acurl "$base" "$token" || true)
  check "workspace list is JSON array" "[" "$resp"

  # 3. Memory commit round-trip — scope=LOCAL so test data stays on this
  # tenant. Verifies encryption + scrubber + retrieval end-to-end.
  probe_id="canary-smoke-$(date +%s)-$i"
  body=$(printf '{"scope":"LOCAL","namespace":"canary-smoke","content":"probe-%s"}' "$probe_id")
-  CANARY_ACURL_PATH="/memories/commit" resp=$(curl -sS --max-time 20 \
+  ACURL_PATH="/memories/commit" resp=$(curl -sS --max-time 20 \
    -X POST -H "Content-Type: application/json" -H "Authorization: Bearer $token" \
    --data "$body" "$base/memories/commit" || true)
  check "memory commit accepted" '"id"' "$resp"

-  CANARY_ACURL_PATH="/memories/search?query=probe-${probe_id}" \
+  ACURL_PATH="/memories/search?query=probe-${probe_id}" \
    resp=$(curl -sS --max-time 20 -H "Authorization: Bearer $token" \
    "$base/memories/search?query=probe-${probe_id}" || true)
  check "memory search finds the probe" "probe-${probe_id}" "$resp"

  # 4. Events admin read — AdminAuth path (C4 fail-closed proof on SaaS).
-  CANARY_ACURL_PATH="/events" resp=$(acurl "$base" "$token" || true)
+  ACURL_PATH="/events" resp=$(acurl "$base" "$token" || true)
  check "events endpoint returns JSON" "[" "$resp"

  # 5. Negative: unauth'd admin call must 401 (C4 regression gate).
@ -117,7 +128,7 @@ for i in "${!URLS[@]}"; do
  # 6. POST /org/import unauth → 401. Proves the route is compiled in
  # and AdminAuth is enforced. A missing route returns 404 (the failure
  # mode caught by issue #213). Regression guard for the silent-GHCR-
-  # migration gap: canary-verify was testing a stale GHCR image while
+  # migration gap: staging-verify (formerly canary-verify) was testing a stale GHCR image while
  # actual tenants ran ECR — this test would have caught a missing-route
  # binary before it reached prod.
  unauth_code=$(curl -sS -o /dev/null -w '%{http_code}' \
--- a/tests/e2e/STAGING_SAAS_E2E.md
+++ b/tests/e2e/STAGING_SAAS_E2E.md
@ -7,11 +7,11 @@ Four workflows + a shared bash harness that together cover the SaaS stack end to
 | Workflow | Cadence | Wall time | Scope |
 |---|---|---|---|
 | `e2e-staging-saas.yml` | push + nightly 07:00 UTC | ~20 min | Full API: org → tenant → 2 workspaces → A2A → HMA → delegation → leak check |
-| `canary-staging.yml` | every 30 min | ~8 min | Minimum smoke + self-managed alert issue |
+| `staging-smoke.yml` | every 30 min | ~8 min | Minimum smoke + self-managed alert issue |
 | `e2e-staging-canvas.yml` | push + weekly Sunday 08:00 | ~25 min | All 13 canvas workspace-panel tabs via Playwright |
 | `e2e-staging-sanity.yml` | weekly Monday 06:00 | ~10 min | Intentional-failure: teardown safety-net self-check |

-`tests/e2e/test_staging_full_saas.sh` is the shared harness all workflows invoke (with `E2E_MODE={full|canary}` and `E2E_INTENTIONAL_FAILURE={0|1}` toggles).
+`tests/e2e/test_staging_full_saas.sh` is the shared harness all workflows invoke (with `E2E_MODE={full|smoke}` and `E2E_INTENTIONAL_FAILURE={0|1}` toggles).

 ### Full-SaaS checklist (sections)

@ -49,7 +49,15 @@ Runs the harness with `E2E_INTENTIONAL_FAILURE=1`, which poisons the tenant admi

 Set in **Settings → Secrets and variables → Actions → Repository secrets**:

-### `MOLECULE_STAGING_ADMIN_TOKEN`
+### `CP_STAGING_ADMIN_API_TOKEN`
+
+> **Historical-rename note (2026-05-11):** previously named
+> `MOLECULE_STAGING_ADMIN_TOKEN`. Canonicalised to
+> `CP_STAGING_ADMIN_API_TOKEN` per internal#322 (the Railway staging
+> service exposes it as `CP_ADMIN_API_TOKEN`; the `CP_*` repo-secret
+> prefix matches the upstream env name + makes the service it talks
+> to obvious in workflow YAMLs). See the original PR for the
+> cross-workflow sweep.

 The `CP_ADMIN_API_TOKEN` env currently set on the Railway staging molecule-platform → controlplane service.

@ -82,7 +90,7 @@ bash tests/e2e/test_staging_full_saas.sh
 ## Cost

 - Full run: ~20 min, ~$0.007
- Canary (48/day): ~$0.06/day
+- Smoke (48/day): ~$0.06/day
 - Canvas (few/week): ~$0.01/day
 - Sanity (weekly): ~$0.002/week
 - **Total staging burn: < $0.15/day** at expected CI load
--- a/tests/e2e/test_staging_full_saas.sh
+++ b/tests/e2e/test_staging_full_saas.sh
@ -27,7 +27,11 @@
 #   E2E_PROVISION_TIMEOUT_SECS   default 900 (15 min cold EC2 budget)
 #   E2E_KEEP_ORG                 1 → skip teardown (debugging only)
 #   E2E_RUN_ID                   Slug suffix; CI: ${GITHUB_RUN_ID}
-#   E2E_MODE                     full (default) | canary
+#   E2E_MODE                     full (default) | smoke
+#                                (legacy alias `canary` still accepted —
+#                                 mapped to `smoke` for back-compat with
+#                                 any in-flight runner picking up an older
+#                                 workflow checkout)
 #   E2E_INTENTIONAL_FAILURE      1 → poison tenant token mid-run so the
 #                                script fails; the EXIT trap MUST still
 #                                tear down cleanly (and exit 4 on leak).
@ -49,15 +53,23 @@ RUNTIME="${E2E_RUNTIME:-hermes}"
 PROVISION_TIMEOUT_SECS="${E2E_PROVISION_TIMEOUT_SECS:-900}"
 RUN_ID_SUFFIX="${E2E_RUN_ID:-$(date +%H%M%S)-$$}"
 MODE="${E2E_MODE:-full}"
+# `canary` is a legacy alias for `smoke` retained for back-compat with
+# any in-flight runner picking up an older workflow checkout during the
+# 2026-05-11 canary→staging rename rollout. Both map to the same slug
+# prefix below. Remove the `canary` alias after one week of no-old-mode
+# observations.
+if [ "$MODE" = "canary" ]; then
+  MODE="smoke"
+fi
 case "$MODE" in
-  full|canary) ;;
-  *) echo "E2E_MODE must be 'full' or 'canary' (got: $MODE)" >&2; exit 2 ;;
+  full|smoke) ;;
+  *) echo "E2E_MODE must be 'full' or 'smoke' (got: $MODE)" >&2; exit 2 ;;
 esac

-# Canary runs get a distinct prefix so their safety-net sweeper only
+# Smoke runs get a distinct slug prefix so their safety-net sweeper only
 # touches their own runs, not in-flight full runs.
-if [ "$MODE" = "canary" ]; then
-  SLUG="e2e-canary-$(date +%Y%m%d)-${RUN_ID_SUFFIX}"
+if [ "$MODE" = "smoke" ]; then
+  SLUG="e2e-smoke-$(date +%Y%m%d)-${RUN_ID_SUFFIX}"
 else
  SLUG="e2e-$(date +%Y%m%d)-${RUN_ID_SUFFIX}"
 fi