jonah-fleet 1.1.0 → 1.4.0

package/templates/prompts/product-planning.md

@@ -2,7 +2,7 @@
 
 ## Objective
 
-Turn strategy and evidence into **well-scoped, ready-to-build work** and a **current roadmap** — the "plan" between analytics/measurement and autowork's build phase. Each run reads roadmap priorities, analytics signals, and user feedback to **propose** the next units of product work: new feature issues, formal specs/PRDs for larger epics, backlog re-ranking recommendations, and roadmap updates.
+Turn strategy and evidence into **well-scoped, ready-to-build work** and a **current roadmap** — the "plan" between analytics/measurement and autowork's build phase. Each run reads roadmap priorities, closed measurement trackers (`RECOMMENDATION: [PIVOT | DEPRECATE | ITERATE]`), analytics signals, and user feedback to **propose** the next units of product work: new feature issues, feature pruning and deprecations, formal specs/PRDs for larger epics, backlog re-ranking recommendations, and roadmap updates.
 
 This routine runs behind a **human approval gate**: it **never files autowork-ready issues on its own**. It stages proposals in one planning issue for operator approval; only a subsequent operator-approved **Promote** run creates the real issues autowork can claim.
 
@@ -11,9 +11,10 @@ This routine runs behind a **human approval gate**: it **never files autowork-re
 This routine runs in two modes: **Propose** (scheduled cron sweep / unapproved fire) and **Promote** (operator-approved fire).
 
 In **Propose mode**, SUCCESS requires:
-- [ ] Read current roadmap, domain documentation, and recent feedback/analytics findings
+- [ ] Read current roadmap, domain documentation, closed measurement trackers, and recent feedback/analytics findings
+- [ ] Performed Feature Pruning & Deprecation Audit: evaluated shipped features and measurement outcomes for low-ROI (<2% user adoption or >50% failure rate) features, drafting deprecation, removal, or simplification proposals
 - [ ] Created or updated exactly one dated staging issue (`🗺️ Product Plan — {date}`) containing:
-  - Up to 3 well-scoped proposals (Summary/Tasks/Why/Complexity)
+  - Up to 3 well-scoped proposals (Summary/Tasks/Why/Complexity), covering additions, pivots, or deprecations
   - Backlog re-ranking recommendations
   - Formal `/to-spec` PRD drafts for any proposal above `size/M`
   - Proposed `ROADMAP.md` updates
@@ -33,6 +34,13 @@ If any criterion cannot be met, stop immediately and log FAILURE with the reason
 - **Max iterations**: 30 — after 30 tool call rounds without completing Definition of Done, STOP. Log FAILURE with category `token_limit`.
 - **Max scope**: planning and staging only. Do not implement code or open feature PRs.
 - **Language Requirement**: All GitHub issue titles, descriptions, task checklists, and comments MUST be written in **English**.
+- **Session link footer**: sign every GitHub post with the Antigravity run footer (`_Generated by [Antigravity](${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID})_`).
+
+## Negative examples (DO NOT do these)
+
+- Do not ignore closed measurement tracker directives (`RECOMMENDATION: [PIVOT | DEPRECATE | ITERATE]`) — incorporate them directly into planning.
+- Do not propose only new features while ignoring dead features with <2% adoption or >50% failure rates — keep the codebase lean by actively proposing deprecations.
+- Do not create unapproved autowork-ready candidate issues directly in Propose mode.
 
 ## Instructions
 
@@ -43,10 +51,13 @@ If any criterion cannot be met, stop immediately and log FAILURE with the reason
 
 ### Steps 1–3: Propose Mode (Staging Proposals)
 
-1. Read `ROADMAP.md`, `AGENTS.md`, and open issues.
-2. Draft up to 3 high-impact proposals based on roadmap priorities and user feedback.
-3. For proposals sized `size/M` or above, draft a formal specification using `/to-spec`.
-4. Stage all proposals in a dedicated staging issue: `🗺️ Product Plan — {YYYY-MM-DD}` assigned to the repo maintainer.
+1. Read `ROADMAP.md`, `AGENTS.md`, closed measurement trackers with `RECOMMENDATION: [PIVOT | DEPRECATE | ITERATE]`, and open issues.
+2. **Feature Pruning & Deprecation Audit**:
+   - Audit shipped features and closed measurement tracker verdicts.
+   - For any feature with <2% user adoption, sub-threshold CTR, or >50% failure rate, draft explicit deprecation, removal, or pivot proposals to keep the codebase lean and eliminate maintenance waste.
+3. Draft up to 3 high-impact proposals (including additions, pivots, or deprecations) based on roadmap priorities, measurement outcomes, and user feedback.
+4. For proposals sized `size/M` or above, draft a formal specification using `/to-spec`.
+5. Stage all proposals in a dedicated staging issue: `🗺️ Product Plan — {YYYY-MM-DD}` assigned to the repo maintainer.
 
 ### Step 4: Promote Mode (Approved Execution)
 

package/templates/skills/diagnosing-bugs/SKILL.md

@@ -85,6 +85,15 @@ Done when **every remaining element is load-bearing**: removing any one of them
 
 Do not proceed until you have reproduced **and** minimised.
 
+## Intent vs. Defect Guardrail (Telemetry Rabbit Hole Prevention)
+
+When diagnosing an under-performing event, low-conversion funnel, or zero-click CTA:
+
+1. **Check for mechanical functionality first.** Run the code path or click handler in isolation: does it execute, route, and emit telemetry without throwing?
+2. **Distinguish user intent from code defects.** If the feature renders above the fold and works when clicked, but telemetry reveals near-zero clicks (<2% adoption), the root cause is **lack of user intent**, not a software bug.
+3. **Prevent the Telemetry Rabbit Hole.** Do NOT spend cycles instrumenting elaborate fallback telemetry, defensive error handling, or retry loops for a feature users simply do not want.
+4. **Handoff to product planning.** Reclassify the issue as a product/UX question (`needs-design` or `roadmap/*`) and hand it off to `product-planning` rather than writing code fixes.
+
 ## Phase 3: Hypothesise
 
 Generate **3–5 ranked hypotheses** before testing any of them. Single-hypothesis generation anchors on the first plausible idea.

package/templates/workflows/autowork-cron.yml

@@ -12,6 +12,10 @@ on:
         description: 'Optional issue number to work in Targeted mode (leave blank for Scan mode)'
         required: false
         type: string
+      issue_number:
+        description: 'Alias for target_issue (e.g. 18)'
+        required: false
+        type: string
 
 concurrency:
   group: autowork-agent-scan
@@ -106,7 +110,7 @@ jobs:
           GOOGLE_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           ANTIGRAVITY_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
-          TARGET_ISSUE: ${{ inputs.target_issue }}
+          TARGET_ISSUE: ${{ inputs.target_issue || inputs.issue_number }}
           SKILLS_PROMPT: ${{ steps.skills.outputs.skills_prompt }}
           GITHUB_SERVER_URL: ${{ github.server_url }}
           GITHUB_REPOSITORY: ${{ github.repository }}
@@ -132,6 +136,20 @@ jobs:
             cat "$LATEST_LOG" >> "$GITHUB_STEP_SUMMARY"
           fi
 
+      - name: Emit Fleet Telemetry (Opt-in)
+        if: always()
+        env:
+          GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
+          TELEMETRY_ENDPOINT: ${{ secrets.JONAH_FLEET_TELEMETRY_ENDPOINT }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          GITHUB_RUN_ID: ${{ github.run_id }}
+          GITHUB_RUN_NUMBER: ${{ github.run_number }}
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/autowork/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ] && [ -f "$LATEST_LOG" ]; then
+            npx --yes jonah-fleet telemetry --emit --log "$LATEST_LOG" --endpoint "$TELEMETRY_ENDPOINT" || true
+          fi
+
       - name: Verify Atomic Handoff Invariant (PR Existence Guard)
         if: always()
         env:

package/templates/workflows/prompt-optimizer-cron.yml

@@ -83,3 +83,25 @@ jobs:
             --output-format text \
             --print-timeout 35m \
             --dangerously-skip-permissions
+
+      - name: Emit run summary
+        if: always()
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/optimizer/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ] && [ -n "$GITHUB_STEP_SUMMARY" ]; then
+            cat "$LATEST_LOG" >> "$GITHUB_STEP_SUMMARY"
+          fi
+
+      - name: Emit Fleet Telemetry (Opt-in)
+        if: always()
+        env:
+          GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
+          TELEMETRY_ENDPOINT: ${{ secrets.JONAH_FLEET_TELEMETRY_ENDPOINT }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          GITHUB_RUN_ID: ${{ github.run_id }}
+          GITHUB_RUN_NUMBER: ${{ github.run_number }}
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/optimizer/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ] && [ -f "$LATEST_LOG" ]; then
+            npx --yes jonah-fleet telemetry --emit --log "$LATEST_LOG" --endpoint "$TELEMETRY_ENDPOINT" || true
+          fi

package/templates/workflows/trigger-autowork-manual.yml

@@ -0,0 +1,169 @@
+name: Trigger Autowork Manually (Targeted)
+
+# Manual execution of the Autowork routine on a specific issue via workflow_dispatch or badge link click.
+
+on:
+  workflow_dispatch:
+    inputs:
+      issue_number:
+        description: 'Target issue number to work in Targeted mode (e.g. 18)'
+        required: true
+        type: string
+      mode:
+        description: 'Execution mode (targeted or scan)'
+        required: false
+        default: 'targeted'
+        type: choice
+        options:
+          - targeted
+          - scan
+      custom_prompt:
+        description: 'Optional custom instructions or overrides'
+        required: false
+        type: string
+
+concurrency:
+  group: autowork-manual-${{ inputs.issue_number || github.run_id }}
+  cancel-in-progress: false
+
+jobs:
+  run-autowork:
+    runs-on: ubuntu-latest
+    timeout-minutes: 65
+    permissions:
+      contents: write
+      issues: write
+      pull-requests: write
+      statuses: write
+      actions: read
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0
+
+      - name: Setup Node.js
+        uses: actions/setup-node@v4
+        with:
+          node-version: 22
+
+      - name: Configure Git author identity
+        run: |
+          git config --global user.name "github-actions[bot]"
+          git config --global user.email "github-actions[bot]@users.noreply.github.com"
+          git config --global --add safe.directory "$GITHUB_WORKSPACE"
+
+      - name: Cache Antigravity CLI
+        id: cache-agy
+        uses: actions/cache@v4
+        with:
+          path: ~/.local/bin/agy
+          key: agy-cli-linux-amd64-v1
+
+      - name: Install Antigravity CLI
+        if: steps.cache-agy.outputs.cache-hit != 'true'
+        run: |
+          curl -fsSL https://antigravity.google/cli/install.sh | bash
+
+      - name: Add Antigravity CLI to PATH
+        run: echo "$HOME/.local/bin" >> $GITHUB_PATH
+
+      - name: Write Antigravity OAuth Token
+        env:
+          ANTIGRAVITY_OAUTH_TOKEN: ${{ secrets.ANTIGRAVITY_OAUTH_TOKEN || secrets.GEMINI_API_KEY }}
+        run: |
+          if [ -z "$ANTIGRAVITY_OAUTH_TOKEN" ]; then
+            echo "::error::Neither ANTIGRAVITY_OAUTH_TOKEN nor GEMINI_API_KEY secret is set; cannot authenticate Antigravity CLI." >&2
+            exit 1
+          fi
+          mkdir -p "$HOME/.gemini/antigravity-cli"
+          chmod 700 "$HOME/.gemini/antigravity-cli"
+          printf '%s' "$ANTIGRAVITY_OAUTH_TOKEN" > "$HOME/.gemini/antigravity-cli/antigravity-oauth-token"
+          chmod 600 "$HOME/.gemini/antigravity-cli/antigravity-oauth-token"
+
+      - name: Preflight validation
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: |
+          if [ ! -s "$HOME/.gemini/antigravity-cli/antigravity-oauth-token" ]; then
+            echo "::error::Preflight check failed: Antigravity OAuth token file is missing or empty." >&2
+            exit 1
+          fi
+          if ! gh auth status >/dev/null 2>&1; then
+            echo "::error::GitHub CLI authentication check failed with GITHUB_TOKEN." >&2
+            exit 1
+          fi
+          echo "✓ Preflight credential and GitHub CLI validation passed."
+
+      - name: Discover domain skills
+        id: skills
+        run: |
+          SKILLS_PROMPT=""
+          if [ -d ".agents/skills" ]; then
+            for skill_path in .agents/skills/*/SKILL.md; do
+              if [ -f "$skill_path" ]; then
+                SKILLS_PROMPT="${SKILLS_PROMPT}Read and follow ${skill_path}. "
+              fi
+            done
+          fi
+          echo "skills_prompt=$SKILLS_PROMPT" >> "$GITHUB_OUTPUT"
+
+      - name: Run Autowork
+        timeout-minutes: 60
+        env:
+          GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+          GOOGLE_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+          ANTIGRAVITY_API_KEY: ${{ secrets.GEMINI_API_KEY }}
+          GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
+          TARGET_ISSUE: ${{ inputs.issue_number }}
+          MODE: ${{ inputs.mode || 'targeted' }}
+          CUSTOM_PROMPT: ${{ inputs.custom_prompt }}
+          SKILLS_PROMPT: ${{ steps.skills.outputs.skills_prompt }}
+          GITHUB_SERVER_URL: ${{ github.server_url }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          GITHUB_RUN_ID: ${{ github.run_id }}
+        run: |
+          if [ -n "$TARGET_ISSUE" ] && [ "$MODE" != "scan" ]; then
+            PROMPT="You are the Autowork routine for this repository. Read and follow the instructions in .github/prompts/autowork.md exactly. ${SKILLS_PROMPT} Your target is issue #${TARGET_ISSUE}. You are in Targeted mode: work issue #${TARGET_ISSUE} directly, ahead of Phase 1 convergence and priority scan."
+          else
+            PROMPT="You are the Autowork routine for this repository. Read and follow the instructions in .github/prompts/autowork.md exactly. ${SKILLS_PROMPT} You are in Scan mode: check open PRs for review comments to fix, close merged issues, then pick the highest-priority unclaimed issue."
+          fi
+
+          if [ -n "$CUSTOM_PROMPT" ]; then
+            PROMPT="${PROMPT} Additional instructions: ${CUSTOM_PROMPT}"
+          fi
+
+          agy -p "$PROMPT" \
+            --model gemini-3.7-flash-high \
+            --output-format text \
+            --print-timeout 60m \
+            --dangerously-skip-permissions
+
+      - name: Emit run summary
+        if: always()
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/autowork/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ] && [ -n "$GITHUB_STEP_SUMMARY" ]; then
+            cat "$LATEST_LOG" >> "$GITHUB_STEP_SUMMARY"
+          fi
+
+      - name: Verify Atomic Handoff Invariant (PR Existence Guard)
+        if: always()
+        env:
+          GH_TOKEN: ${{ github.token }}
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/autowork/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ]; then
+            if grep -q "Result.*SUCCESS" "$LATEST_LOG"; then
+              WORKED_BRANCH=$(grep -E '^\s*-\s*(Pushed branch|Branch):' "$LATEST_LOG" | head -n 1 | sed -E 's/.*`([^`]+)`.*/\1/')
+              if [ -n "$WORKED_BRANCH" ] && [[ "$WORKED_BRANCH" =~ (feat|fix)/ ]]; then
+                echo "Verifying PR existence for branch: $WORKED_BRANCH"
+                PR_COUNT=$(gh pr list --head "$WORKED_BRANCH" --state all --json number --jq 'length' 2>/dev/null || echo "0")
+                if [ "$PR_COUNT" -eq 0 ]; then
+                  echo "::error::Atomic Handoff Invariant Violated: Branch '$WORKED_BRANCH' was implemented and pushed, but no pull request exists on GitHub!" >&2
+                  exit 1
+                fi
+                echo "✓ Verified pull request exists on GitHub for branch: $WORKED_BRANCH"
+              fi
+            fi
+          fi

package/templates/workflows/trigger-autowork-on-bug.yml

@@ -11,10 +11,10 @@ concurrency:
 jobs:
   fire-autowork:
     if: >-
-      (github.event.action == 'opened' && contains(github.event.issue.labels.*.name, 'bug') && contains(github.event.issue.labels.*.name, 'priority/P1')) ||
-      (github.event.action == 'labeled' && (github.event.label.name == 'bug' || github.event.label.name == 'priority/P1'))
+      (github.event.action == 'opened' && contains(github.event.issue.labels.*.name, 'bug') && (contains(github.event.issue.labels.*.name, 'priority/P1') || contains(github.event.issue.labels.*.name, 'priority/P0'))) ||
+      (github.event.action == 'labeled' && contains(github.event.issue.labels.*.name, 'bug') && (contains(github.event.issue.labels.*.name, 'priority/P1') || contains(github.event.issue.labels.*.name, 'priority/P0')))
     runs-on: ubuntu-latest
-    timeout-minutes: 65
+    timeout-minutes: 35
     permissions:
       contents: write
       issues: write
@@ -80,7 +80,7 @@ jobs:
           echo "skills_prompt=$SKILLS_PROMPT" >> "$GITHUB_OUTPUT"
 
       - name: Run Autowork
-        timeout-minutes: 60
+        timeout-minutes: 30
         env:
           GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           GOOGLE_API_KEY: ${{ secrets.GEMINI_API_KEY }}
@@ -97,5 +97,5 @@ jobs:
           agy -p "$PROMPT" \
             --model gemini-3.7-flash-high \
             --output-format text \
-            --print-timeout 60m \
+            --print-timeout 30m \
             --dangerously-skip-permissions

package/templates/workflows/trigger-autowork-on-merge.yml

@@ -15,7 +15,7 @@ jobs:
       github.event.pull_request.base.ref == 'main' &&
       github.event.pull_request.head.repo.full_name == github.repository
     runs-on: ubuntu-latest
-    timeout-minutes: 65
+    timeout-minutes: 35
     permissions:
       contents: write
       issues: write
@@ -110,7 +110,7 @@ jobs:
 
       - name: Run Autowork
         if: steps.pick.outputs.skip != 'true'
-        timeout-minutes: 60
+        timeout-minutes: 30
         env:
           GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           GOOGLE_API_KEY: ${{ secrets.GEMINI_API_KEY }}
@@ -126,5 +126,5 @@ jobs:
           agy -p "$PROMPT" \
             --model gemini-3.7-flash-high \
             --output-format text \
-            --print-timeout 60m \
+            --print-timeout 30m \
             --dangerously-skip-permissions

package/templates/workflows/trigger-review-routine.yml

@@ -2,21 +2,48 @@ name: Trigger Review Routine
 
 on:
   pull_request:
-    types: [ready_for_review, opened, reopened, synchronize]
+    types: [ready_for_review, opened, reopened, synchronize, review_requested]
     paths-ignore:
       - '.github/prompts/logs/**'
+  schedule:
+    # Periodic scan sweep every 2 hours to catch and recover any orphaned or unreviewed ready PRs
+    - cron: '45 */2 * * *'
+  workflow_dispatch:
+    inputs:
+      pr_number:
+        description: 'Optional PR number to review in Targeted mode (leave blank for Scan mode)'
+        required: false
+        type: string
+  issue_comment:
+    types: [created]
 
 concurrency:
-  group: review-routine-pr-${{ github.event.pull_request.number }}
+  group: review-routine-pr-${{ github.event.pull_request.number || github.event.issue.number || inputs.pr_number || 'scan' }}
   cancel-in-progress: true
 
 jobs:
   trigger-routine:
     if: >-
-      github.event.pull_request.draft == false &&
-      github.event.pull_request.head.repo.full_name == github.repository
+      (github.event_name == 'pull_request' &&
+       github.event.pull_request.draft == false &&
+       github.event.pull_request.head.repo.full_name == github.repository &&
+       !startsWith(github.event.pull_request.head.ref, 'release-please--')) ||
+      (github.event_name == 'schedule') ||
+      (github.event_name == 'workflow_dispatch') ||
+      (github.event_name == 'issue_comment' &&
+       github.event.issue.pull_request != null &&
+       !endsWith(github.event.comment.user.login, '[bot]') &&
+       github.event.comment.user.type != 'Bot' &&
+       (startsWith(github.event.comment.body, '/review') ||
+        startsWith(github.event.comment.body, '/peer-review') ||
+        startsWith(github.event.comment.body, '/retrigger') ||
+        startsWith(github.event.comment.body, '/re-review') ||
+        contains(github.event.comment.body, '/review') ||
+        contains(github.event.comment.body, '/peer-review') ||
+        contains(github.event.comment.body, '/retrigger') ||
+        contains(github.event.comment.body, '/re-review')))
     runs-on: ubuntu-latest
-    timeout-minutes: 60
+    timeout-minutes: 35
     permissions:
       contents: write
       actions: read
@@ -28,7 +55,21 @@ jobs:
         id: guard
         env:
           GH_TOKEN: ${{ github.token }}
+          PR_NUM: ${{ github.event.pull_request.number || github.event.issue.number || inputs.pr_number }}
         run: |
+          if [ -z "$PR_NUM" ]; then
+            if [ "${{ github.event_name }}" = "schedule" ]; then
+              open_prs=$(gh api "repos/${{ github.repository }}/pulls?state=open&per_page=1" --jq 'length' 2>/dev/null || echo "0")
+              if [ "$open_prs" -eq "0" ]; then
+                echo "No open PRs found in ${{ github.repository }} — skipping scheduled review routine scan."
+                echo "skip=true" >> "$GITHUB_OUTPUT"
+                exit 0
+              fi
+            fi
+            echo "skip=false" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+
           gh_api_retry() {
             local endpoint="$1"
             shift
@@ -46,18 +87,37 @@ jobs:
             return 1
           }
 
-          if ! pr=$(gh_api_retry "repos/${{ github.repository }}/pulls/${{ github.event.pull_request.number }}"); then
+          if ! pr=$(gh_api_retry "repos/${{ github.repository }}/pulls/${PR_NUM}"); then
             echo "skip=false" >> "$GITHUB_OUTPUT"
             exit 0
           fi
 
           state=$(echo "$pr" | jq -r '.state')
           merged=$(echo "$pr" | jq -r '.merged')
+          head_ref=$(echo "$pr" | jq -r '.head.ref')
           if [ "$state" = "closed" ] || [ "$merged" = "true" ]; then
             echo "PR is already $state/merged=$merged — skipping review."
             echo "skip=true" >> "$GITHUB_OUTPUT"
             exit 0
           fi
+          if [[ "$head_ref" == release-please--* ]]; then
+            echo "PR head branch $head_ref is an automated release PR — skipping review."
+            echo "skip=true" >> "$GITHUB_OUTPUT"
+            exit 0
+          fi
+
+          # Priority-driven dual execution filter on direct pull_request triggers:
+          # If PR is labeled priority/P2 or priority/P3, skip immediate cloud review to let local agent daemons review it
+          # (cloud scheduled watchdog sweep will catch up if unreviewed > 48h).
+          if [ "${{ github.event_name }}" = "pull_request" ]; then
+            labels=$(echo "$pr" | jq -r '.labels[].name' 2>/dev/null || echo "")
+            if echo "$labels" | grep -qE '^priority/P[23]$'; then
+              echo "PR is labeled as lower priority (P2/P3) — skipping immediate cloud review for local peer-review daemon."
+              echo "skip=true" >> "$GITHUB_OUTPUT"
+              exit 0
+            fi
+          fi
+
           echo "skip=false" >> "$GITHUB_OUTPUT"
 
       - name: Checkout repository
@@ -112,26 +172,48 @@ jobs:
 
       - name: Run Peer Review
         if: steps.guard.outputs.skip != 'true'
-        timeout-minutes: 55
+        timeout-minutes: 30
         env:
           GEMINI_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           GOOGLE_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           ANTIGRAVITY_API_KEY: ${{ secrets.GEMINI_API_KEY }}
           GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
-          PR_NUMBER: ${{ github.event.pull_request.number }}
-          PR_URL: ${{ github.event.pull_request.html_url }}
+          PR_NUMBER: ${{ github.event.pull_request.number || github.event.issue.number || inputs.pr_number }}
+          PR_URL: ${{ github.event.pull_request.html_url || github.event.issue.html_url }}
           GITHUB_SERVER_URL: ${{ github.server_url }}
           GITHUB_REPOSITORY: ${{ github.repository }}
           GITHUB_RUN_ID: ${{ github.run_id }}
         run: |
-          PROMPT="You are the Peer Review routine for this repository. Read and follow the instructions in .github/prompts/peer-review.md exactly. Your target is pull request #${PR_NUMBER} (${PR_URL}). You are in Targeted mode: review PR #${PR_NUMBER} directly."
+          if [ -n "$PR_NUMBER" ]; then
+            if [ -z "$PR_URL" ] || [[ "$PR_URL" == *"/issues/"* ]]; then
+              PR_URL="${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/pull/${PR_NUMBER}"
+            fi
+            PROMPT="You are the Peer Review routine for this repository. Read and follow the instructions in .github/prompts/peer-review.md exactly. Your target is pull request #${PR_NUMBER} (${PR_URL}). You are in Targeted mode: review PR #${PR_NUMBER} directly."
+          else
+            PROMPT="You are the Peer Review routine for this repository. Read and follow the instructions in .github/prompts/peer-review.md exactly. You are in Scan mode: check open PRs and select the highest-priority PR to review."
+          fi
 
           agy -p "$PROMPT" \
             --model gemini-3.7-flash-high \
             --output-format text \
-            --print-timeout 55m \
+            --print-timeout 30m \
             --dangerously-skip-permissions
 
+      - name: Handle Review Failure & PR Notification
+        if: failure() && steps.guard.outputs.skip != 'true'
+        env:
+          GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
+          PR_NUMBER: ${{ github.event.pull_request.number || github.event.issue.number || inputs.pr_number }}
+          GITHUB_SERVER_URL: ${{ github.server_url }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          GITHUB_RUN_ID: ${{ github.run_id }}
+        run: |
+          if [ -n "$PR_NUMBER" ]; then
+            gh pr comment "$PR_NUMBER" --body "⚠️ **Peer Review Routine Notice**: The automated review session encountered an execution error (e.g., API rate/quota limit or runner interruption). The review will be retried automatically on the next scheduled scan sweep or can be re-triggered immediately by commenting \`/review\` on this PR.
+
+          _Generated by [Antigravity](${GITHUB_SERVER_URL}/${GITHUB_REPOSITORY}/actions/runs/${GITHUB_RUN_ID})_" || true
+          fi
+
       - name: Emit run summary
         if: always() && steps.guard.outputs.skip != 'true'
         run: |
@@ -139,3 +221,17 @@ jobs:
           if [ -n "$LATEST_LOG" ] && [ -n "$GITHUB_STEP_SUMMARY" ]; then
             cat "$LATEST_LOG" >> "$GITHUB_STEP_SUMMARY"
           fi
+
+      - name: Emit Fleet Telemetry (Opt-in)
+        if: always() && steps.guard.outputs.skip != 'true'
+        env:
+          GH_TOKEN: ${{ secrets.GH_PAT || github.token }}
+          TELEMETRY_ENDPOINT: ${{ secrets.JONAH_FLEET_TELEMETRY_ENDPOINT }}
+          GITHUB_REPOSITORY: ${{ github.repository }}
+          GITHUB_RUN_ID: ${{ github.run_id }}
+          GITHUB_RUN_NUMBER: ${{ github.run_number }}
+        run: |
+          LATEST_LOG=$(ls -t .github/prompts/logs/peer-review/*.md 2>/dev/null | head -n 1)
+          if [ -n "$LATEST_LOG" ] && [ -f "$LATEST_LOG" ]; then
+            npx --yes jonah-fleet telemetry --emit --log "$LATEST_LOG" --endpoint "$TELEMETRY_ENDPOINT" || true
+          fi