context-engine-ai/Context Engine Reviewer
AI-powered PR reviews for Context Engine workflows.
View on GitHubTrust Signals
- Scorecard Score
- not yet scored
- Maintenance Recency
- Stale
- License
- None
Inputs
| name | description | required | default |
|---|---|---|---|
| style_guide_rules | Custom style guide rules that will be enforced during review by generating critical comments | no | "" |
| github_api_url | GitHub API URL for GitHub Enterprise Server (e.g., https://github.example.com/api/v3) | no | https://api.github.com |
| github_server_url | GitHub Server URL for GitHub Enterprise Server (e.g., https://github.example.com) | no | https://github.com |
| custom_mode | Enhanced senior engineer review mode: on | off | auto | no | auto |
| llm_model | Override LLM model (e.g., gpt-4o-mini) | no | "" |
| llm_provider | Override LLM provider (e.g., ai-sdk) | no | "" |
| llm_reasoning_effort | Reasoning effort for models that support it (e.g., low | medium | high). Claude Fable is always pinned to medium. | no | "" |
| zai_base_url | Z.AI OpenAI-compatible base URL for glm-* models | no | https://api.z.ai/api/coding/paas/v4/ |
| kimi_base_url | Kimi (Moonshot AI) OpenAI-compatible base URL for kimi-* and moonshot-* models | no | https://api.moonshot.ai/v1 |
| deepseek_base_url | DeepSeek OpenAI-compatible base URL for deepseek-* models | no | https://api.deepseek.com/v1 |
| context_engine_api_key | Optional Context Engine API key. When set, reviewer LLMs can call Context Engine MCP tools for repository context. | no | "" |
| context_engine_mcp_url | Optional Context Engine MCP indexer URL. | no | https://dev.context-engine.ai/indexer/mcp |
| context_engine_collection | Optional Context Engine collection name to scope reviewer MCP searches. | no | "" |
| context_engine_tools | Optional comma-separated Context Engine MCP tools exposed to the reviewer LLM. | no | repo_search,batch_search,symbol_graph,batch_symbol_graph,graph_query,batch_graph_query,search_tests_for,search_config_for,search_commits_for,context_answer |
| context_engine_max_tools | Maximum number of Context Engine MCP tools exposed to the reviewer LLM. | no | 10 |
| context_engine_max_steps | Maximum LLM steps (tool roundtrips plus final output) per review batch when Context Engine tools are enabled. | no | 8 |
| review_scopes | Comma-separated review scopes to run: security,performance,best-practices | no | "" |
| max_comments | Maximum inline comments to post per run | no | 40 |
| max_codeblock_lines | Maximum lines retained inside fenced code blocks in comments | no | 60 |
| max_review_chars | Approximate maximum diff characters per LLM review batch | no | 725000 |
| allow_title_update | Allow the action to update PR title when @context-engine-reviewer is present (default: false) | no | false |
Outputs
no outputs