diff --git a/.github/workflows/pr-agent.yml b/.github/workflows/pr-agent.yml
new file mode 100644
index 00000000..c704c2a0
--- /dev/null
+++ b/.github/workflows/pr-agent.yml
@@ -0,0 +1,30 @@
+name: PR Agent (DeepSeek)
+
+on:
+  pull_request:
+    types: [opened, reopened, ready_for_review]
+  issue_comment:
+    types: [created]
+
+jobs:
+  pr_agent_job:
+    if: ${{ github.event.sender.type != 'Bot' }}
+    runs-on: ubuntu-latest
+    permissions:
+      issues: write
+      pull-requests: write
+      contents: write
+    name: Run PR Agent on every pull request, respond to user comments
+    steps:
+      - name: PR Agent action step
+        id: pragent
+        uses: the-pr-agent/pr-agent@main
+        env:
+          OPENAI_KEY: ${{ secrets.DEEPSEEK_API_KEY }}
+          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+          CONFIG.MODEL: "deepseek/deepseek-chat"
+          CONFIG.FALLBACK_MODELS: '["deepseek/deepseek-chat"]'
+          GITHUB_ACTION_CONFIG.AUTO_REVIEW: "true"
+          GITHUB_ACTION_CONFIG.AUTO_DESCRIBE: "true"
+          GITHUB_ACTION_CONFIG.AUTO_IMPROVE: "true"
+          CONFIG.AI_TIMEOUT: "180"
diff --git a/.pr_agent.toml b/.pr_agent.toml
new file mode 100644
index 00000000..2309109f
--- /dev/null
+++ b/.pr_agent.toml
@@ -0,0 +1,36 @@
+[config]
+model = "deepseek/deepseek-chat"
+fallback_models = ["deepseek/deepseek-chat"]
+ai_timeout = 180
+
+[github_action_config]
+auto_review = true
+auto_describe = true
+auto_improve = true
+pr_actions = ["opened", "reopened", "ready_for_review"]
+
+[pr_reviewer]
+extra_instructions = """\
+Focus on the following aspects during review:
+- Security vulnerabilities and unsafe code patterns
+- Logic errors and edge cases not handled
+- Performance bottlenecks and inefficient algorithms
+- API design, breaking changes, and backward compatibility
+- Proper error handling and logging
+- Code maintainability, readability, and adherence to project conventions
+- Missing tests or insufficient test coverage for the changes
+"""
+num_max_findings = 5
+require_tests_review = true
+require_security_review = true
+require_estimate_effort_to_review = true
+
+[pr_description]
+generate_ai_title = true
+use_bullet_points = true
+enable_pr_diagram = true
+
+[pr_code_suggestions]
+num_code_suggestions = 6
+suggestions_score_threshold = 5
+focus_only_on_problems = true
diff --git a/README.md b/README.md
index 26e7fccd..02e930a9 100644
--- a/README.md
+++ b/README.md
@@ -128,4 +128,6 @@ dotnet test --collect:"XPlat Code Coverage"
 See the canonical AI rule set under `docs/ai/rules/`:
 - **Rule Index**: `docs/ai/rules/README.md`
 - **Copilot Entrypoint**: `.github/copilot-instructions.md`
-- **Agent Entrypoint**: `AGENTS.md`
\ No newline at end of file
+- **Agent Entrypoint**: `AGENTS.md`
+
+<!-- PR-Agent verification -->
\ No newline at end of file