Pattern miner: Add cross-repository monitor archetype from contribution-check pattern

.github/copilot-instructions.md

@@ -4,7 +4,7 @@ You are an expert GitHub Agentic Workflow generator. You help users create produ
 
 ## Your Knowledge
 
-Use the committed pattern library as the source of truth: `patterns/manifest.json` plus `patterns/archetypes/*.json` generated on 2026-09-28 from 563 source repos, 345 active workflows, and 929 total workflows scanned. The current wizard manifest lists 33 user-facing archetypes; the `custom` archetype exists as a supporting pattern file and is intentionally not exposed as a HOW-step archetype.
+Use the committed pattern library as the source of truth: `patterns/manifest.json` plus `patterns/archetypes/*.json` generated on 2026-09-28 from 563 source repos, 345 active workflows, and 929 total workflows scanned. The current wizard manifest lists 34 user-facing archetypes; the `custom` archetype exists as a supporting pattern file and is intentionally not exposed as a HOW-step archetype.
 
 ### Key Data Points
 
@@ -24,7 +24,7 @@ Use the committed pattern library as the source of truth: `patterns/manifest.jso
 - `custom` is hidden from the wizard archetype cards but retained for matching and profile data. Best observed custom profiles are schedule + create-pull-request + noop at 95.2% (n=21), schedule + create-issue + noop + threat-detection at 83.9%, and schedule + create-issue + noop at 80.0%.
 
 **Curated archetypes without empirical runs yet (`count: 0`):**
-- accessibility-expert, agent-cost-tracker, backlog-drip, batched-ci-doctor, ci-failure-triage, code-health-auditor, community-digest, contribution-guidelines-checker, issue-hierarchy-manager, link-checker, linter-applier, linter-miner, linter-refiner, linter-workflows, nitpick-reviewer, performance-nut, pr-fix-assistant, pr-iteration-loop, repo-qa-assistant, responsive-docs-tester, security-scanner, user-simulator, vex-statement-generator.
+- accessibility-expert, agent-cost-tracker, backlog-drip, batched-ci-doctor, ci-failure-triage, code-health-auditor, community-digest, contribution-guidelines-checker, cross-repo-monitor, issue-hierarchy-manager, link-checker, linter-applier, linter-miner, linter-refiner, linter-workflows, nitpick-reviewer, performance-nut, pr-fix-assistant, pr-iteration-loop, repo-qa-assistant, responsive-docs-tester, security-scanner, user-simulator, vex-statement-generator.
 - Keep these archetypes available. They are newer curated patterns and should not be removed simply because they have no measured success rate.
 
 **Trigger combo risk:**

patterns/archetypes/cross-repo-monitor.json

@@ -0,0 +1,39 @@
+{
+  "id": "cross-repo-monitor",
+  "label": "Cross-Repository Monitor",
+  "description": "Read and evaluate activity in a different target repository on a schedule, dispatch per-item evaluation, and write a consolidated report plus labels/comments back to the target repo",
+  "success_rate": null,
+  "count": 0,
+  "recommended_triggers": [
+    {
+      "type": "schedule",
+      "config": {}
+    }
+  ],
+  "recommended_safe_outputs": [
+    "issues",
+    "add-labels",
+    "add-comment"
+  ],
+  "recommended_tools": [
+    "create-issue",
+    "add-labels",
+    "add-comment"
+  ],
+  "timeout_minutes": 20,
+  "prompt_style": "role-steps",
+  "size_range_bytes": [
+    3000,
+    8000
+  ],
+  "top_repos": [],
+  "tips": [
+    "Set target-repo (an env var or repository variable, not a hardcoded string) on every safe output that writes to the external repository, and read from that same repository via GitHub toolsets",
+    "Pre-filter and fetch the target repository's items in a steps: block before the agent starts, so the agent only evaluates a bounded, pre-computed list",
+    "Act as an orchestrator: dispatch each item to a stateless, self-contained evaluation (inline sub-agent or per-item prompt) and compile the results into one report rather than evaluating everything inline",
+    "Publish the consolidated report as an issue in the home repository, and use close-older-issues: true so each run replaces the previous report instead of accumulating",
+    "Use add-labels and add-comment with target-repo to apply verdicts and feedback directly on items in the target repository, each capped with max: to bound write volume",
+    "Document the required token scopes (contents, issues, pull-requests, or actions:write for cross-repo dispatch) in the prompt since the default GITHUB_TOKEN cannot write to another repository"
+  ],
+  "anti_patterns": []
+}

patterns/archetypes/issue-hierarchy-manager.json

@@ -27,7 +27,8 @@
     "Pre-fetch open issues in a steps: block before reasoning about relationships",
     "Use link-sub-issue with parent/sub label or title-prefix filters to bound which issues can be linked",
     "Recursively close parent issues only after confirming every sub-issue is complete",
-    "Read-only issues permission is enough \u2014 rely on safe-outputs like link-sub-issue and update-issue for writes"
+    "Read-only issues permission is enough \u2014 rely on safe-outputs like link-sub-issue and update-issue for writes",
+    "When this run needs to create several new findings as a set, use create-issue's group: true so they land as linked sub-issues under one parent instead of unrelated top-level issues"
   ],
   "anti_patterns": []
 }

patterns/manifest.json

@@ -38,7 +38,8 @@
     "pr-fix-assistant",
     "backlog-drip",
     "responsive-docs-tester",
-    "vex-statement-generator"
+    "vex-statement-generator",
+    "cross-repo-monitor"
   ],
   "workflow_generation": "workflow-generation.json",
   "anti_patterns": [

patterns/workflow-generation.json

@@ -544,6 +544,41 @@
         "- **DO NOT** report a finding already covered by an existing open issue from this workflow."
       ]
     },
+    "cross-repo-monitor": {
+      "icon": "repo",
+      "capabilities": {
+        "pre_steps": true,
+        "bash": true,
+        "github_toolsets": true
+      },
+      "permissions": [
+        "contents",
+        "issues",
+        "pull-requests"
+      ],
+      "instructions": [
+        "workflow-patterns.md"
+      ],
+      "body": [
+        "# {{label}}",
+        "",
+        "You are an **orchestrator** that monitors a different target repository and reports findings back to this repository.",
+        "",
+        "{{pre_steps}}",
+        "## Process",
+        "",
+        "1. Read the pre-filtered list of items from the target repository produced in the steps above",
+        "2. Dispatch each item to a stateless, self-contained evaluation and collect its verdict",
+        "3. Compile all verdicts into a single consolidated report issue in THIS repository",
+        "4. Apply labels and post comments on the evaluated items in the target repository using the pre-configured `target-repo` safe outputs",
+        "",
+        "## Constraints",
+        "",
+        "- **DO NOT** write to the target repository with anything other than the configured safe-output tools \u2014 never use `gh` or the GitHub API directly for target-repo writes.",
+        "- **DO NOT** evaluate more items than the pre-filtered list provides.",
+        "- **DO NOT** merge, approve, or close items in the target repository."
+      ]
+    },
     "vex-statement-generator": {
       "icon": "shield",
       "capabilities": {