ci: incrementally update docs graph on PR merge

.github/workflows/update-graph.yml

@@ -0,0 +1,115 @@
+# Incrementally updates the FalkorDB docs knowledge graph on PR merge.
+#
+# On merge to main of a PR touching .md files, this workflow:
+#   1. Computes the diff (added / modified / deleted .md files).
+#   2. Reads the content of added + modified files.
+#   3. POSTs the payload to GraphRAG-UI's /api/admin/update-graph endpoint.
+#   4. The endpoint does all the SDK / FalkorDB / smoke-test work server-side
+#      using its existing credentials. This workflow only needs ONE secret:
+#      ``UPDATE_GRAPH_TOKEN`` — a shared-secret bearer token.
+#
+# No Azure OpenAI keys, no FalkorDB credentials, no PAT for cross-repo
+# checkout. Failures from the server come back as HTTP non-2xx with a
+# detail message; curl --fail-with-body bubbles them up as a CI failure.
+
+name: Update graph (incremental)
+
+on:
+  pull_request:
+    types: [closed]
+    branches: [main]
+    paths:
+      - "**/*.md"
+
+# Merge target is always main, so all runs share one group. Bursts of PR
+# merges queue rather than race. cancel-in-progress: false because each
+# run consumes LLM credit and we'd rather pay the wait than re-do work.
+concurrency:
+  group: update-graph-${{ github.ref }}
+  cancel-in-progress: false
+
+jobs:
+  update-graph:
+    if: github.event.pull_request.merged == true
+    runs-on: ubuntu-latest
+    timeout-minutes: 30
+    env:
+      GRAPH_ID: docs_benchmark
+      # The base URL of the GraphRAG-UI deployment. Set this as a repo or
+      # environment variable so it can differ between staging and prod.
+      GRAPHRAG_UI_URL: ${{ vars.GRAPHRAG_UI_URL }}
+    steps:
+      - name: Checkout docs
+        uses: actions/checkout@v4
+        with:
+          fetch-depth: 0  # need history for diff against the PR base
+
+      - name: Build diff payload
+        id: payload
+        env:
+          BASE_SHA: ${{ github.event.pull_request.base.sha }}
+          HEAD_SHA: ${{ github.sha }}
+        run: |
+          python3 - <<'PY'
+          import json, os, pathlib, subprocess, sys
+
+          base, head = os.environ["BASE_SHA"], os.environ["HEAD_SHA"]
+          out = subprocess.run(
+              ["git", "diff", "--name-status", base, head],
+              capture_output=True, text=True, check=True,
+          ).stdout
+
+          added, modified, deleted = {}, {}, []
+          for line in out.splitlines():
+              parts = line.split("\t")
+              if not parts:
+                  continue
+              status = parts[0][0]  # R100 -> R
+              if status == "R" and len(parts) >= 3:
+                  old, new = parts[1], parts[2]
+                  if old.endswith(".md"):
+                      deleted.append(old)
+                  if new.endswith(".md"):
+                      try:
+                          added[new] = pathlib.Path(new).read_text(encoding="utf-8")
+                      except FileNotFoundError:
+                          pass
+                  continue
+              if len(parts) < 2 or not parts[1].endswith(".md"):
+                  continue
+              path = parts[1]
+              if status == "A":
+                  added[path] = pathlib.Path(path).read_text(encoding="utf-8")
+              elif status == "M":
+                  modified[path] = pathlib.Path(path).read_text(encoding="utf-8")
+              elif status == "D":
+                  deleted.append(path)
+
+          payload = {
+              "graph_id": os.environ.get("GRAPH_ID", "docs_benchmark"),
+              "files": {"added": added, "modified": modified, "deleted": deleted},
+          }
+          if not (added or modified or deleted):
+              # Path filter on the workflow trigger should make this rare, but
+              # be explicit: no work to do, exit clean before the POST.
+              print("::notice::No .md changes after filtering — skipping graph update.", file=sys.stderr)
+              with open(os.environ["GITHUB_OUTPUT"], "a") as f:
+                  f.write("skip=true\n")
+              sys.exit(0)
+
+          pathlib.Path("payload.json").write_text(json.dumps(payload))
+          print(f"::notice::Diff: +{len(added)} ~{len(modified)} -{len(deleted)} files")
+          with open(os.environ["GITHUB_OUTPUT"], "a") as f:
+              f.write("skip=false\n")
+          PY
+
+      - name: Call admin update-graph endpoint
+        if: steps.payload.outputs.skip != 'true'
+        run: |
+          curl -X POST "$GRAPHRAG_UI_URL/api/admin/update-graph" \
+            -H "Authorization: Bearer ${{ secrets.UPDATE_GRAPH_TOKEN }}" \
+            -H "Content-Type: application/json" \
+            --data-binary @payload.json \
+            --fail-with-body \
+            --show-error \
+            --max-time 1800