codereview
Read-only
Performs systematic, step-by-step code review covering quality, security, performance, and architecture with expert validation.
Instructions
Performs systematic, step-by-step code review with expert validation. Use for comprehensive analysis covering quality, security, performance, and architecture. Guides through structured investigation to ensure thoroughness.
Input Schema
TableJSON Schema
| Name | Required | Description | Default |
|---|---|---|---|
| step | Yes | Review narrative. Step 1: outline the review strategy. Later steps: report findings. MUST cover quality, security, performance, and architecture. Reference code via `relevant_files`; avoid dumping large snippets. | |
| model | Yes | Currently in auto model selection mode. CRITICAL: When the user names a model, you MUST use that exact name unless the server rejects it. If no model is provided, you may use the `listmodels` tool to review options and select an appropriate match. Top models: gemini-2.5-pro (score 100, 1.0M ctx, thinking, code-gen); gemini-3-pro-preview (score 100, 1.0M ctx, thinking, code-gen); gemini-2.5-flash (score 61, 1.0M ctx, thinking); gemini-2.0-flash (score 56, 1.0M ctx, thinking); gemini-2.0-flash-lite (score 42, 1.0M ctx). | |
| images | No | Optional diagram or screenshot paths that clarify review context. | |
| findings | Yes | Capture findings (positive and negative) across quality, security, performance, and architecture; update each step. | |
| focus_on | No | Optional note on areas to emphasise (e.g. 'threading', 'auth flow'). | |
| standards | No | Coding standards or style guides to enforce. | |
| confidence | No | Confidence level: exploring (just starting), low (early investigation), medium (some evidence), high (strong evidence), very_high (comprehensive understanding), almost_certain (near complete confidence), certain (100% confidence locally - no external validation needed) | |
| hypothesis | No | Current theory about issue/goal based on work | |
| review_type | No | Review focus: full, security, performance, or quick. | full |
| step_number | Yes | Current review step (starts at 1) – each step should build on the last. | |
| temperature | No | 0 = deterministic · 1 = creative. | |
| total_steps | Yes | Number of review steps planned. External validation: two steps (analysis + summary). Internal validation: one step. Use the same limits when continuing an existing review via continuation_id. | |
| issues_found | No | Issues with severity (critical/high/medium/low) and descriptions. | |
| files_checked | No | Absolute paths of every file reviewed, including those ruled out. | |
| thinking_mode | No | Reasoning depth: minimal, low, medium, high, or max. | |
| relevant_files | No | Step 1: list all files/dirs under review. Must be absolute full non-abbreviated paths. Final step: narrow to files tied to key findings. | |
| continuation_id | No | Unique thread continuation ID for multi-turn conversations. Works across different tools. ALWAYS reuse the last continuation_id you were given—this preserves full conversation context, files, and findings so the agent can resume seamlessly. | |
| severity_filter | No | Lowest severity to include when reporting issues (critical/high/medium/low/all). | all |
| relevant_context | No | Methods/functions identified as involved in the issue | |
| next_step_required | Yes | True when another review step follows. External validation: step 1 → True, step 2 → False. Internal validation: set False immediately. Apply the same rule on continuation flows. | |
| use_assistant_model | No | Use assistant model for expert analysis after workflow steps. False skips expert analysis, relies solely on your personal investigation. Defaults to True for comprehensive validation. | |
| review_validation_type | No | Set 'external' (default) for expert follow-up or 'internal' for local-only review. | external |