From bdbecb40278b10dddbe628428d37e3b28efeebf0 Mon Sep 17 00:00:00 2001
From: CyberWolf <haywolf@qq.com>
Date: Wed, 4 Nov 2026 12:40:31 +0800
Subject: [PATCH] chore: remove pr-review workflow

---
 .github/workflows/pr-review.yml | 117 --------------------------------
 1 file changed, 117 deletions(-)
 delete mode 100644 .github/workflows/pr-review.yml

diff --git a/.github/workflows/pr-review.yml b/.github/workflows/pr-review.yml
deleted file mode 100644
index 0ab3625..0000000
--- a/.github/workflows/pr-review.yml
+++ /dev/null
@@ -1,117 +0,0 @@
-name: Code Reviewer Workflow
-
-on:
-  pull_request:
-    types: [opened, synchronize, reopened, ready_for_review]
-
-permissions:
-  contents: read
-  pull-requests: write
-  issues: write
-
-jobs:
-  review:
-    runs-on: ubuntu-latest
-    env:
-      LLM_API_KEY: ${{ secrets.LLM_API_KEY }}
-    steps:
-      - name: Checkout Current Branch
-        uses: actions/checkout@v4
-
-      - name: Check LLM_API_KEY
-        if: ${{ env.LLM_API_KEY == '' }}
-        run: echo "LLM_API_KEY is not set. Skip self-test run."
-
-      - name: Run Action From Current Branch
-        if: ${{ env.LLM_API_KEY != '' }}
-        uses: TiyAgents/code-review-agent-action@v3
-        with:
-          github_token: ${{ secrets.GITHUB_TOKEN }}
-          # Provider 配置（新字段）
-          ai_provider: ${{ vars.LLM_PROVIDER_PROTOCOL }}
-          api_key: ${{ env.LLM_API_KEY }}
-          api_base: ${{ vars.LLM_API_BASE }}
-
-          # [可选] Planner 使用模型；默认 gpt-5.3-codex
-          # 可选值：任意可用模型名（字符串）
-          planner_model: ${{ vars.PLANNER_MODEL }}
-
-          # [可选] SubAgent 使用模型；默认 gpt-5.3-codex
-          # 可选值：任意可用模型名（字符串）
-          reviewer_model: ${{ vars.DEFAULT_MODEL }}
-
-          api_base_allowlist: |
-            api.openai.com
-            zenmux.ai
-            opencode.ai
-            openrouter.ai
-            tokenhub.tencentmaas.com
-            api.lkeap.cloud.tencent.com
-
-          # [Optional] Include globs (comma or newline separated); default is ** (all files)
-          # Allowed values: any valid glob patterns
-          include: |
-            **/*.js
-            **/*.ts
-            **/*.tsx
-            **/*.json
-
-          # [Optional] Exclude globs (comma or newline separated); default is empty
-          # Allowed values: any valid glob patterns
-          exclude: |
-            **/*.lock
-            **/dist/**
-            **/*.min.js
-            **/release/**
-            **/node_modules/**
-            **/*.log
-
-          # [Optional] Review dimensions (comma or newline separated);
-          # default is general,security,performance,testing
-          # Suggested values: general, security, performance, testing
-          review_dimensions: general,security,performance,testing
-
-          # [Optional] Review language; default is English
-          # Allowed values: any language label (for example English, Chinese, Simplified Chinese)
-          review_language: English
-
-          # [Optional] Minimum confidence to keep a finding; default is 0.72
-          # Allowed range: 0.0 - 1.0 (float)
-          min_finding_confidence: 0.85
-
-          # [Optional] Round-1 coverage-first mode; default is true
-          # Allowed values: true, false
-          # true: first round runs only primary dimension to reduce duplicates/cost
-          coverage_first_round_primary_only: true
-
-          # [Optional] Auto-minimize outdated historical inline comments from this action; default is true
-          # Allowed values: true, false
-          auto_minimize_outdated_comments: true
-
-          # [Optional] Maximum review rounds; default is 8
-          # Allowed range: positive integer (>=1)
-          # Rough sizing: increase with max_model_calls if many files are changed
-          max_rounds: 4
-
-          # [Optional] Maximum model calls; default is 40
-          # Allowed range: positive integer (>=1)
-          # Rough sizing: calls ~= rounds * (1 + ceil(patch_files / batch_size) * dimensions)
-          # Example: ~100 changed files with default dimensions usually needs 80-120 calls
-          max_model_calls: 64
-
-          # [Optional] Maximum files per batch; default is 8
-          # Allowed range: positive integer (>=1)
-          # Larger value reduces batch count (and model calls) but increases per-call context size
-          max_files_per_batch: 16
-
-          # [Optional] Maximum context characters per batch; default is 128000
-          # Allowed range: positive integer (>=1)
-          max_context_chars: 256000
-
-          # [Optional] Maximum findings kept after aggregation; default is 60
-          # Allowed range: positive integer (>=1)
-          max_findings: 64
-
-          # [Optional] Maximum inline comments posted in PR review; default is 30
-          # Allowed range: positive integer (>=1)
-          max_inline_comments: 32